jm + disks   8

Two traps in iostat: %util and svctm
Marc Brooker:
As a measure of general IO busyness %util is fairly handy, but as an indication of how much the system is doing compared to what it can do, it's terrible. Iostat's svctm has even fewer redeeming strengths. It's just extremely misleading for most modern storage systems and workloads. Both of these fields are likely to mislead more than inform on modern SSD-based storage systems, and their use should be treated with extreme care.
ioutil  iostat  svctm  ops  ssd  disks  hardware  metrics  stats  linux 
july 2014 by jm
Migrating from MongoDB to Cassandra
Interesting side-effect of using LUKS for full-disk encryption: 'For every disk read, we were pulling in 3MB of data (RA is sectors, SSZ is sector size, 6144*512=3145728 bytes) into cache. Oops. Not only were we doing tons of extra work, but we were trashing our page cache too. The default for the device-mapper used by LUKS under Ubuntu 12.04LTS is incredibly sub-optimal for database usage, especially our usage of Cassandra (more small random reads vs. large rows). We turned this down to 128 sectors — 64KB.'
cassandra  luks  raid  linux  tuning  ops  blockdev  disks  sdd 
february 2014 by jm
Backblaze Blog » What Hard Drive Should I Buy?
Because Backblaze has a history of openness, many readers expected more details in my previous posts. They asked what drive models work best and which last the longest. Given our experience with over 25,000 drives, they asked which ones are good enough that we would buy them again. In this post, I’ll answer those questions.
backblaze  backup  hardware  hdds  storage  disks  ops  via:fanf 
january 2014 by jm
Rasmus' home NAS design
I'm trying to avoid doing this in order to avoid more power consumption and unpopular hardware in the house -- but if necessary, this is a good up-to-date homebuild design
nas  hardware  home  storage  ops  disks 
november 2013 by jm
'Mythbusting Modern Hardware to gain "Mechanical Sympathy"' [slides]
Martin Thompson's latest talk -- taking a few common concepts about modern hardware performance and debunking/confirming them, mythbusters-style
mythbusters  hardware  mechanical-sympathy  martin-thompson  java  performance  cpu  disks  ssd 
may 2013 by jm
AWS forum post on interpreting iostat output for EBS
Great post from AndrewC@EBS on interpreting iostat output on EBS volumes -- from 2009, but still looks reasonable enough
iostat  ebs  disks  hardware  aws  ops 
may 2013 by jm
First 5 Minutes Troubleshooting A Server
quite a good checklist of first steps for troubleshooting. Worth bookmarking for "dstat --top-io --top-bio" alone, which is an absolutely excellent tool and new to me
dstat  server  io  disks  hardware  performance  linux  sysadmin  ops  troubleshooting  checklists  root-cause 
march 2013 by jm
Jeff Dean's list of "Numbers Everyone Should Know"
from a 2007 Google all-hands, the list of typical latency timings from ranging from an L1 cache reference (0.5 nanoseconds) to a CA->NL->CA IP round trip (150 milliseconds).
performance  latencies  google  jeff-dean  timing  caches  speed  network  zippy  disks  via:kellabyte 
march 2013 by jm

Copy this bookmark:



description:


tags: