Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Support
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
M
margo
Project overview
Project overview
Details
Activity
Releases
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Issues
17
Issues
17
List
Boards
Labels
Milestones
Merge Requests
2
Merge Requests
2
Analytics
Analytics
Repository
Value Stream
Wiki
Wiki
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Create a new issue
Commits
Issue Boards
Open sidebar
sds
margo
Commits
f5cc48c9
Commit
f5cc48c9
authored
May 22, 2017
by
Philip Carns
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
add brief readme
parent
c0901e58
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
54 additions
and
0 deletions
+54
-0
examples/composition/README.txt
examples/composition/README.txt
+54
-0
No files found.
examples/composition/README.txt
0 → 100644
View file @
f5cc48c9
The examples in this directory benchmark the overhead of relaying to RPCs to co-located
services in various configurations.
<NOTE: ignore performance results in these examples; they were taken from a debugging build>
The following example will compare the latency of issuing RPCs (with 8 MiB data transfers)
directly to a service vs the latency of the same RPC relayed through a delegator service.
The two daemon instances do not need to run on the same node, but you need to use another
transport like CCI, BMI, or OFI to relay between remote nodes.
==========================
$ ./composed-svc-daemon na+sm:// data-xfer
# accepting RPCs on address "na+sm://24151/0"
$ ./composed-svc-daemon na+sm:// delegator
# accepting RPCs on address "na+sm://24180/0"
$ ./composed-benchmark na+sm://24180/0 na+sm://24151/0 100
# DBG: starting data_xfer_read() benchmark.
# DBG: ... DONE.
# <op> <min> <avg> <max>
direct 0.001103 0.001333 0.005824
# DBG: starting composed_read() benchmark.
# DBG: ... DONE.
# <op> <min> <avg> <max>
composed 0.001103 0.001751 0.005824
# DBG: ... DONE.
Shutting down delegator server.
Shutting down data_xfer server.
==========================
Alternatively, you can run the two services in the same executable, in which case the relay
will bypass the network layer of Mercury entirely.
==========================
$ ./composed-svc-daemon na+sm:// data-xfer,delegator
# accepting RPCs on address "na+sm://24259/0"
$ ./composed-benchmark na+sm://24259/0 na+sm://24259/0 100
# DBG: starting data_xfer_read() benchmark.
# DBG: ... DONE.
# <op> <min> <avg> <max>
direct 0.001215 0.001514 0.010304
# DBG: starting composed_read() benchmark.
# DBG: ... DONE.
# <op> <min> <avg> <max>
composed 0.001215 0.001756 0.010304
# DBG: ... DONE.
Shutting down delegator server.
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment