2
2
mirror of https://github.com/octoleo/restic.git synced 2025-02-13 17:28:27 +00:00
greatroar 5141228e0c repository: Re-tune indexmap allocation strategy
fd05037e1ad7350d89ae91fdb9f52f954787bc4a changed the allocation batch
size from 256 to 128 under the assumption that an indexEntry is 60 bytes
on amd64, but it's 64: structs are padded out to a multiple of 8 for
alignment reasons. That means we'd waste no space in malloc even without
the batch allocation, at least on 64-bit machines. While that strategy
cuts the overallocation down dramatically for many small indexes, it also
seems to slow allocation down (Go 1.18, Linux, amd64, -benchtime=2s):

    name                   old time/op    new time/op    delta
    DecodeIndex-8             4.67s ± 5%     4.60s ± 1%      ~     (p=0.953 n=10+5)
    DecodeIndexParallel-8     4.67s ± 3%     4.60s ± 1%      ~     (p=0.953 n=10+5)
    IndexHasUnknown-8        37.8ns ± 8%    36.5ns ±14%      ~     (p=0.841 n=5+5)
    IndexHasKnown-8          38.5ns ±12%    37.7ns ±10%      ~     (p=0.968 n=5+5)
    IndexAlloc-8              615ms ±18%     607ms ± 1%      ~     (p=1.000 n=10+5)
    IndexAllocParallel-8      245ms ±11%     285ms ± 6%   +16.40%  (p=0.001 n=10+5)
    MasterIndexAlloc-8        286ms ± 9%     275ms ± 2%      ~     (p=1.000 n=10+5)
    LoadIndex/v1-8           27.0ms ± 4%    26.8ms ± 1%      ~     (p=0.690 n=5+5)
    LoadIndex/v2-8           22.4ms ± 1%    22.8ms ± 2%    +1.48%  (p=0.016 n=5+5)

    name                   old alloc/op   new alloc/op   delta
    IndexAlloc-8              446MB ± 0%     446MB ± 0%    -0.00%  (p=0.000 n=8+4)
    IndexAllocParallel-8      446MB ± 0%     446MB ± 0%    -0.00%  (p=0.008 n=8+5)
    MasterIndexAlloc-8        213MB ± 0%     159MB ± 0%   -25.47%  (p=0.000 n=10+5)

    name                   old allocs/op  new allocs/op  delta
    IndexAlloc-8               913k ± 0%     2632k ± 0%  +188.19%  (p=0.008 n=5+5)
    IndexAllocParallel-8       913k ± 0%     2632k ± 0%  +188.21%  (p=0.008 n=5+5)
    MasterIndexAlloc-8         318k ± 0%     1172k ± 0%  +267.86%  (p=0.008 n=5+5)

Instead, this patch sets a batch size of 4, which means no space is
wasted by malloc on 64-bit and very little on 32-bit. It still gets very
close to the savings from not allocating in batches, without requiring
special code for bits.UintSize==64. Benchmark results, again for
Linux/amd64:

    name                   old time/op    new time/op    delta
    DecodeIndex-8             4.67s ± 5%     4.83s ± 9%     ~     (p=0.315 n=10+10)
    DecodeIndexParallel-8     4.67s ± 3%     4.68s ± 4%     ~     (p=0.315 n=10+10)
    IndexHasUnknown-8        37.8ns ± 8%    44.5ns ±19%     ~     (p=0.095 n=5+5)
    IndexHasKnown-8          38.5ns ±12%    36.9ns ± 8%     ~     (p=0.690 n=5+5)
    IndexAlloc-8              615ms ±18%     628ms ±18%     ~     (p=0.218 n=10+10)
    IndexAllocParallel-8      245ms ±11%     262ms ± 9%   +7.02%  (p=0.043 n=10+10)
    MasterIndexAlloc-8        286ms ± 9%     287ms ±13%     ~     (p=1.000 n=10+10)
    LoadIndex/v1-8           27.0ms ± 4%    26.8ms ± 0%     ~     (p=1.000 n=5+5)
    LoadIndex/v2-8           22.4ms ± 1%    22.5ms ± 0%     ~     (p=0.056 n=5+5)

    name                   old alloc/op   new alloc/op   delta
    IndexAlloc-8              446MB ± 0%     446MB ± 0%     ~     (p=1.000 n=8+10)
    IndexAllocParallel-8      446MB ± 0%     446MB ± 0%   -0.00%  (p=0.000 n=8+8)
    MasterIndexAlloc-8        213MB ± 0%     160MB ± 0%  -25.02%  (p=0.000 n=10+9)

    name                   old allocs/op  new allocs/op  delta
    IndexAlloc-8               913k ± 0%     1333k ± 0%  +45.94%  (p=0.000 n=8+10)
    IndexAllocParallel-8       913k ± 0%     1333k ± 0%  +45.94%  (p=0.000 n=8+8)
    MasterIndexAlloc-8         318k ± 0%      525k ± 0%  +64.99%  (p=0.000 n=10+10)

The allocation method indexmap.newEntry has also been rewritten in a
form that is a few instructions shorter.
2022-05-11 21:22:14 +02:00
2022-03-30 21:11:17 +02:00
2021-09-21 08:23:33 +02:00
2021-09-21 08:23:33 +02:00
2020-11-05 09:40:56 +01:00
2022-03-26 20:09:40 +01:00
2022-03-30 21:11:17 +02:00
2022-04-30 20:03:21 +02:00
2022-04-30 20:16:09 +02:00
2018-05-28 22:29:06 +02:00
2017-11-02 11:39:49 +01:00
2020-03-01 11:30:02 +01:00
2022-04-11 20:34:14 +02:00

Documentation Build Status Go Report Card

Introduction

restic is a backup program that is fast, efficient and secure. It supports the three major operating systems (Linux, macOS, Windows) and a few smaller ones (FreeBSD, OpenBSD).

For detailed usage and installation instructions check out the documentation.

You can ask questions in our Discourse forum.

Quick start

Once you've installed restic, start off with creating a repository for your backups:

$ restic init --repo /tmp/backup
enter password for new backend:
enter password again:
created restic backend 085b3c76b9 at /tmp/backup
Please note that knowledge of your password is required to access the repository.
Losing your password means that your data is irrecoverably lost.

and add some data:

$ restic --repo /tmp/backup backup ~/work
enter password for repository:
scan [/home/user/work]
scanned 764 directories, 1816 files in 0:00
[0:29] 100.00%  54.732 MiB/s  1.582 GiB / 1.582 GiB  2580 / 2580 items  0 errors  ETA 0:00
duration: 0:29, 54.47MiB/s
snapshot 40dc1520 saved

Next you can either use restic restore to restore files or use restic mount to mount the repository via fuse and browse the files from previous snapshots.

For more options check out the online documentation.

Backends

Saving a backup on the same machine is nice but not a real backup strategy. Therefore, restic supports the following backends for storing backups natively:

Design Principles

Restic is a program that does backups right and was designed with the following principles in mind:

  • Easy: Doing backups should be a frictionless process, otherwise you might be tempted to skip it. Restic should be easy to configure and use, so that, in the event of a data loss, you can just restore it. Likewise, restoring data should not be complicated.

  • Fast: Backing up your data with restic should only be limited by your network or hard disk bandwidth so that you can backup your files every day. Nobody does backups if it takes too much time. Restoring backups should only transfer data that is needed for the files that are to be restored, so that this process is also fast.

  • Verifiable: Much more important than backup is restore, so restic enables you to easily verify that all data can be restored.

  • Secure: Restic uses cryptography to guarantee confidentiality and integrity of your data. The location the backup data is stored is assumed not to be a trusted environment (e.g. a shared space where others like system administrators are able to access your backups). Restic is built to secure your data against such attackers.

  • Efficient: With the growth of data, additional snapshots should only take the storage of the actual increment. Even more, duplicate data should be de-duplicated before it is actually written to the storage back end to save precious backup space.

Reproducible Builds

The binaries released with each restic version starting at 0.6.1 are reproducible, which means that you can reproduce a byte identical version from the source code for that release. Instructions on how to do that are contained in the builder repository.

News

You can follow the restic project on Twitter @resticbackup or by subscribing to the project blog.

License

Restic is licensed under BSD 2-Clause License. You can find the complete text in LICENSE.

Sponsorship

Backend integration tests for Google Cloud Storage and Microsoft Azure Blob Storage are sponsored by AppsCode!

Sponsored by AppsCode

Description
No description provided
Readme BSD-2-Clause 66 MiB
Languages
Go 99.7%
Shell 0.3%