Patchwork worker: document poor partitioning scheme impact

login
register
mail settings
Submitter Gregory Szorc
Date Feb. 28, 2016, 5:43 a.m.
Message ID <0a50d8a09039b16c9293.1456638206@ubuntu-vm-main>
Download mbox | patch
Permalink /patch/13446/
State Accepted
Headers show

Comments

Gregory Szorc - Feb. 28, 2016, 5:43 a.m.
# HG changeset patch
# User Gregory Szorc <gregory.szorc@gmail.com>
# Date 1456638197 28800
#      Sat Feb 27 21:43:17 2016 -0800
# Node ID 0a50d8a09039b16c9293bff4ec32bf414ca35710
# Parent  41dcd754526612c43b9695df8851557c851828ef
worker: document poor partitioning scheme impact

mpm isn't a fan of the existing or previous partitioning scheme. He
provided a fantastic justification for why on the mailing list.

This patch adds his words to the code so they aren't forgotten.

Patch

diff --git a/mercurial/worker.py b/mercurial/worker.py
--- a/mercurial/worker.py
+++ b/mercurial/worker.py
@@ -152,11 +152,33 @@  if os.name != 'nt':
     _exitstatus = _posixexitstatus
 
 def partition(lst, nslices):
     '''partition a list into N slices of roughly equal size
 
     The current strategy takes every Nth element from the input. If
     we ever write workers that need to preserve grouping in input
     we should consider allowing callers to specify a partition strategy.
+
+    mpm is not a fan of this partitioning strategy when files are involved.
+    In his words:
+
+        Single-threaded Mercurial makes a point of creating and visiting
+        files in a fixed order (alphabetical). When creating files in order,
+        a typical filesystem is likely to allocate them on nearby regions on
+        disk. Thus, when revisiting in the same order, locality is maximized
+        and various forms of OS and disk-level caching and read-ahead get a
+        chance to work.
+
+        This effect can be quite significant on spinning disks. I discovered it
+        circa Mercurial v0.4 when revlogs were named by hashes of filenames.
+        Tarring a repo and copying it to another disk effectively randomized
+        the revlog ordering on disk by sorting the revlogs by hash and suddenly
+        performance of my kernel checkout benchmark dropped by ~10x because the
+        "working set" of sectors visited no longer fit in the drive's cache and
+        the workload switched from streaming to random I/O.
+
+        What we should really be doing is have workers read filenames from a
+        ordered queue. This preserves locality and also keeps any worker from
+        getting more than one file out of balance.
     '''
     for i in range(nslices):
         yield lst[i::nslices]