s3fs-fuse/src/fdcache.h

190 lines
6.1 KiB
C
Raw Normal View History

2014-09-07 15:08:27 +00:00
/*
* s3fs - FUSE-based file system backed by Amazon S3
*
* Copyright 2007-2008 Randy Rizun <rrizun@gmail.com>
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
#ifndef FD_CACHE_H_
#define FD_CACHE_H_
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
//------------------------------------------------
// CacheFileStat
//------------------------------------------------
class CacheFileStat
{
private:
std::string path;
int fd;
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
private:
static bool MakeCacheFileStatPath(const char* path, std::string& sfile_path, bool is_create_dir = true);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
public:
static bool DeleteCacheFileStat(const char* path);
static bool CheckCacheFileStatTopDir(void);
explicit CacheFileStat(const char* tpath = NULL);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
~CacheFileStat();
bool Open(void);
bool Release(void);
bool SetPath(const char* tpath, bool is_open = true);
int GetFd(void) const { return fd; }
};
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
//------------------------------------------------
// fdpage & PageList
//------------------------------------------------
// page block information
struct fdpage
{
off_t offset;
size_t bytes;
bool init;
fdpage(off_t start = 0, size_t size = 0, bool is_init = false)
: offset(start), bytes(size), init(is_init) {}
off_t next(void) const { return (offset + bytes); }
off_t end(void) const { return (0 < bytes ? offset + bytes - 1 : 0); }
};
typedef std::list<struct fdpage*> fdpage_list_t;
//
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
// Management of loading area/modifying
//
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
class PageList
{
private:
fdpage_list_t pages;
private:
void Clear(void);
public:
static void FreeList(fdpage_list_t& list);
explicit PageList(off_t size = 0, bool is_init = false);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
~PageList();
off_t Size(void) const;
int Resize(off_t size, bool is_init);
int Init(off_t size, bool is_init);
bool IsInit(off_t start, off_t size);
bool SetInit(off_t start, off_t size, bool is_init = true);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
bool FindUninitPage(off_t start, off_t& resstart, size_t& ressize);
int GetUninitPages(fdpage_list_t& uninit_list, off_t start = 0, off_t size = -1);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
bool Serialize(CacheFileStat& file, bool is_output);
void Dump(void);
};
//------------------------------------------------
// class FdEntity
//------------------------------------------------
class FdEntity
{
private:
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
pthread_mutex_t fdent_lock;
bool is_lock_init;
PageList pagelist;
int refcnt; // reference count
std::string path; // object path
std::string cachepath; // local cache file path
2015-07-10 18:50:40 +00:00
int fd; // file descriptor(tmp file or cache file)
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
FILE* file; // file pointer(tmp file or cache file)
bool is_modify; // if file is changed, this flag is true
private:
void Clear(void);
int Dup(void);
bool SetAllStatus(bool is_enable);
public:
explicit FdEntity(const char* tpath = NULL, const char* cpath = NULL);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
~FdEntity();
void Close(void);
bool IsOpen(void) const { return (-1 != fd); }
int Open(off_t size = -1, time_t time = -1);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
const char* GetPath(void) const { return path.c_str(); }
void SetPath(const std::string &newpath) { path = newpath; }
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
int GetFd(void) const { return fd; }
int SetMtime(time_t time);
bool GetSize(off_t& size);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
bool GetMtime(time_t& time);
bool GetStats(struct stat& st);
bool SetAllEnable(void) { return SetAllStatus(true); }
bool SetAllDisable(void) { return SetAllStatus(false); }
bool LoadFull(off_t* size = NULL, bool force_load = false);
int Load(off_t start, off_t size);
2014-08-26 17:11:10 +00:00
int RowFlush(const char* tpath, headers_t& meta, bool force_sync = false);
int Flush(headers_t& meta, bool force_sync = false) { return RowFlush(NULL, meta, force_sync); }
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
ssize_t Read(char* bytes, off_t start, size_t size, bool force_load = false);
ssize_t Write(const char* bytes, off_t start, size_t size);
};
typedef std::map<std::string, class FdEntity*> fdent_map_t; // key=path, value=FdEntity*
//------------------------------------------------
// class FdManager
//------------------------------------------------
class FdManager
{
private:
static FdManager singleton;
static pthread_mutex_t fd_manager_lock;
static bool is_lock_init;
static std::string cache_dir;
static size_t page_size;
fdent_map_t fent;
public:
FdManager();
~FdManager();
// Reference singleton
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
static FdManager* get(void) { return &singleton; }
static bool DeleteCacheDirectory(void);
static int DeleteCacheFile(const char* path);
static bool SetCacheDir(const char* dir);
static bool IsCacheDir(void) { return (0 < FdManager::cache_dir.size()); }
static const char* GetCacheDir(void) { return FdManager::cache_dir.c_str(); }
static size_t SetPageSize(size_t size);
static size_t GetPageSize(void) { return FdManager::page_size; }
static bool MakeCachePath(const char* path, std::string& cache_path, bool is_create_dir = true);
static bool CheckCacheTopDir(void);
static bool MakeRandomTempPath(const char* path, std::string& tmppath);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
FdEntity* GetFdEntity(const char* path, int existfd = -1);
FdEntity* Open(const char* path, off_t size = -1, time_t time = -1, bool force_tmpfile = false, bool is_create = true);
FdEntity* ExistOpen(const char* path, int existfd = -1);
void Rename(const std::string &from, const std::string &to);
Changes codes for performance(part 3) * Summay This revision includes big change about temporary file and local cache file. By this big change, s3fs works with good performance when s3fs opens/ closes/syncs/reads object. I made a big change about the handling about temporary file and local cache file to do this implementation. * Detail 1) About temporary file(local file) s3fs uses a temporary file on local file system when s3fs does download/ upload/open/seek object on S3. After this revision, s3fs calls ftruncate() function when s3fs makes the temporary file. In this way s3fs can set a file size of precisely length without downloading. (Notice - ftruncate function is for XSI-compliant systems, so that possibly you have a problem on non-XSI-compliant systems.) By this change, s3fs can download a part of a object by requesting with "Range" http header. It seems like downloading by each block unit. The default block(part) size is 50MB, it is caused the result which is default parallel requests count(5) by default multipart upload size(10MB). If you need to change this block size, you can change by new option "fd_page_size". This option can take from 1MB(1024 * 1024) to any bytes. So that, you have to take care about that fdcache.cpp(and fdcache.h) were changed a lot. 2) About local cache Local cache files which are in directory specified by "use_cache" option do not have always all of object data. This cause is that s3fs uses ftruncate function and reads(writes) each block unit of a temporary file. s3fs manages each block unit's status which are "downloaded area" or "not". For this status, s3fs makes new temporary file in cache directory which is specified by "use_cache" option. This status files is in a directory which is named "<use_cache sirectory>/.<bucket_name>/". When s3fs opens this status file, s3fs locks this file for exclusive control by calling flock function. You need to take care about this, the status files can not be laid on network drive(like NFS). This revision changes about file open mode, s3fs always opens a local cache file and each status file with writable mode. Last, this revision adds new option "del_cache", this option means that s3fs deletes all local cache file when s3fs starts and exits. 3) Uploading When s3fs writes data to file descriptor through FUSE request, old s3fs revision downloads all of the object. But new revision does not download all, it downloads only small percial area(some block units) including writing data area. And when s3fs closes or flushes the file descriptor, s3fs downloads other area which is not downloaded from server. After that, s3fs uploads all of data. Already r456 revision has parallel upload function, then this revision with r456 and r457 are very big change for performance. 4) Downloading By changing a temporary file and a local cache file, when s3fs downloads a object, it downloads only the required range(some block units). And s3fs downloads units by parallel GET request, it is same as a case of uploading. (Maximum parallel request count and each download size are specified same parameters for uploading.) In the new revision, when s3fs opens file, s3fs returns file descriptor soon. Because s3fs only opens(makes) the file descriptor with no downloading data. And when s3fs reads a data, s3fs downloads only some block unit including specified area. This result is good for performance. 5) Changes option name The option "parallel_upload" which added at r456 is changed to new option name as "parallel_count". This reason is this option value is not only used by uploading object, but a uploading object also uses this option. (For a while, you can use old option name "parallel_upload" for compatibility.) git-svn-id: http://s3fs.googlecode.com/svn/trunk@458 df820570-a93a-0410-bd06-b72b767a4274
2013-07-23 16:01:48 +00:00
bool Close(FdEntity* ent);
};
#endif // FD_CACHE_H_
2014-09-07 15:08:27 +00:00
/*
* Local variables:
* tab-width: 4
* c-basic-offset: 4
* End:
* vim600: noet sw=4 ts=4 fdm=marker
* vim<600: noet sw=4 ts=4
*/