Aeolus: Performance

From Pmgwiki

Revision as of 23:17, 8 February 2011

Data to be collected for SOSP paper

1. Preliminaries
   1. Configuration questions
      1. Which machine should be used? Eigenharp and Badger are the fastest. Farm best for when lots of machines needed
      2. Is the Authority Service on the same machine that runs the benchmarks?
         1. ok for micro benchmarks (because different processor cores will be used, but big monster should really be distributed
         2. also test to see if it makes a different
      3. Is the Audit Trail Logging Service on the same machine that runs the benchmarks?
         1. turn off logging for microbenchmarks
      4. Is the File Service on the same machine that runs the benchmarks?
   2. (P) indicates that the benchmark is mentioned in the current paper
   3. List is mostly based on Winnie's thesis
   4. Vicky's thesis also computes "with and without logging"
2. Communication costs
   1. inter VN communication (RPC)
   2. intra VN communicaion (i.e., inter thread/aeolus process)
3. Forks and calls
   1. (P) normal fork vs aeolus fork, af with public pid, af with a different pid
   2. (P) normal call vs aeolus call, also aeolus call with public pid, aeolus call with different pid
   3. Note that "normal" calls and forks are not measured in the paper
   4. (P) closure call
   5. rpc call
   6. create, put, get
4. Shared Memory
   1. Shared Objects: (P) put, get (skip create)
   2. Shared Queues: enqueue, waitanddequeue: basic + ipc (skip create)
   3. Shared Locks: Lock, unlock
5. Monster.com: non-aeolus vs. aeolus
6. Skip the following due to rare use
   1. Creation: Normal process creation vs VN creation
   2. Boxes
7. Skip the following due to monster example
   1. (P) Online Store
   2. (P) Secure Wiki
   3. Web: base web service vs. aeolus web service
8. Skip the following: not too interesting
   1. File operations
      1. (P) create dir
      2. (P) create file
      3. (P) list dir
      4. (P) remove file
      5. (P) remove dir
      6. read various size files
      7. write various size files
      8. file stream open and read, various size files
      9. file stream write, various size files
      10. Note: the paper mentions some measurement information about accessing file content
   2. Authority Server