[Bugs] [Bug 1512447] New: [disperse] Keep stripe in in-memory cache for the non aligned write

Mon Nov 13 09:24:44 UTC 2017

https://bugzilla.redhat.com/show_bug.cgi?id=1512447

            Bug ID: 1512447
           Summary: [disperse] Keep stripe in in-memory cache for the non
                    aligned write
           Product: GlusterFS
           Version: 3.13
         Component: disperse
          Severity: medium
          Priority: high
          Assignee: bugs at gluster.org
          Reporter: aspandey at redhat.com
                CC: bugs at gluster.org, pkarampu at redhat.com,
                    sheggodu at redhat.com
        Depends On: 1471753

+++ This bug was initially created as a clone of Bug #1471753 +++

Description of problem:

    Problem:
    Consider an EC volume with configuration  4 + 2.
    The stripe size for this would be 512 * 4 = 2048.
    That means, 2048 bytes of user data stored in one
    stripe. Let's say 2048 + 512 = 2560 bytes are
    already written on this volume. 512 Bytes would
    be in second stripe. Now, if there are sequential
    writes with offset 2560 and of size 1 Byte, we have
    to read the whole stripe, encode it with 1 Byte and
    then again have to write it back. Next, write with
    offset 2561 and size of 1 Byte will again
    READ-MODIFY-WRITE the whole stripe. This is causing
    bad performance because of lots of READ request
    travelling over the network.

    There are some tools and scenario's where such kind
    of load is coming and users are not aware of that.
    Example: fio and zip

    Solution:
    One possible solution to deal with this issue is to
    keep last stripe in memory. This way, we need not to
    read it again and we can save READ fop going over the
    network. Considering the above example, we have to
    keep last 2048 bytes (maximum) in memory per file.

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1.
2.
3.

Actual results:

Expected results:

Additional info:

--- Additional comment from Worker Ant on 2017-07-17 08:24:29 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#1) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-07-19 01:39:03 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#2) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-07-23 04:47:53 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#3) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-08-21 01:31:04 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#4) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-09-18 06:45:22 EDT ---

REVIEW: https://review.gluster.org/17789 (Problem: Consider an EC volume with
configuration  4 + 2. The stripe size for this would be 512 * 4 = 2048. That
means, 2048 bytes of user data stored in one stripe. Let's say 2048 + 512 =
2560 bytes are already written on this volume. 512 Bytes would be in second
stripe. Now, if there are sequential writes with offset 2560 and of size 1
Byte, we have to read the whole stripe, encode it with 1 Byte and then again
have to write it back. Next, write with offset 2561 and size of 1 Byte will
again READ-MODIFY-WRITE the whole stripe. This is causing bad performance
because of lots of READ request travelling over the network.) posted (#5) for
review on master by Ashish Pandey (aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-09-18 06:55:48 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#6) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-09-22 04:26:54 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#7) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-10-03 08:35:56 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#8) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-10-03 13:51:33 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#9) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-10-04 03:51:44 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#10) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-10-05 06:30:10 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#11) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-10-09 13:20:46 EDT ---

REVIEW: https://review.gluster.org/17789 (cluster/ec: Keep last written strip
in in-memory cache) posted (#12) for review on master by Ashish Pandey
(aspandey at redhat.com)

--- Additional comment from Worker Ant on 2017-11-10 17:16:13 EST ---

COMMIT: https://review.gluster.org/17789 committed in master by  

------------- cluster/ec: Keep last written strip in in-memory cache

Problem:
Consider an EC volume with configuration  4 + 2.
The stripe size for this would be 512 * 4 = 2048.
That means, 2048 bytes of user data stored in one
stripe. Let's say 2048 + 512 = 2560 bytes are
already written on this volume. 512 Bytes would
be in second stripe. Now, if there are sequential
writes with offset 2560 and of size 1 Byte, we have
to read the whole stripe, encode it with 1 Byte and
then again have to write it back. Next, write with
offset 2561 and size of 1 Byte will again
READ-MODIFY-WRITE the whole stripe. This is causing
bad performance because of lots of READ request
travelling over the network.

There are some tools and scenario's where such kind
of load is coming and users are not aware of that.
Example: fio and zip

Solution:
One possible solution to deal with this issue is to
keep last stripe in memory. This way, we need not to
read it again and we can save READ fop going over the
network. Considering the above example, we have to
keep last 2048 bytes (maximum) in memory per file.

Change-Id: I3f95e6fc3ff81953646d374c445a40c6886b0b85
BUG: 1471753
Signed-off-by: Ashish Pandey <aspandey at redhat.com>

Referenced Bugs:

https://bugzilla.redhat.com/show_bug.cgi?id=1471753
[Bug 1471753] [disperse] Keep stripe in in-memory cache for the non aligned
write
-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.