Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

performance problem with remote machine in same network #20

Open
Cphilo opened this issue Mar 5, 2016 · 5 comments
Open

performance problem with remote machine in same network #20

Cphilo opened this issue Mar 5, 2016 · 5 comments

Comments

@Cphilo
Copy link

Cphilo commented Mar 5, 2016

I use the benchmark.py to test the performance. My local machine ip is 192.168.0.201 and I have another redis-server running on 192.168.0.101. Here is the parameter with the benchmark.py file.

count    = 50000
capacity = 1e9
error    = 1e-6

Then the benchmark result with 192.168.0.201 is as follows:

Generating 100000 random test words
Generated random test words in 1.082357s
Filter using 20 hash functions and 28755175137 bits
Batch insert : 11.030547s (4532.866814 words / second)
Serial insert: 14.348155s (3484.768594 words / second)
Batch test   : 2.664293s (18766.704355 words / second)
Serial test  : 5.634645s (8873.673520 words / second)
False positive rate: 0.000000 (0.000001 expected)
Redis set add  : 0.119999s (416670.375433 words / second)
Redis pipe chk : 0.884489s (56529.826959 words / second)
Redis pipe sadd: 0.888721s (56260.626908 words / second)
Redis pipe chk : 0.884423s (56534.032935 words / second)

And the benchmark with 192.168.0.101 is as follows:

Generating 100000 random test words
Generated random test words in 1.091547s
Filter using 20 hash functions and 28755175137 bits
Batch insert : 14.148405s (3533.967238 words / second)
Serial insert: 30.745577s (1626.250183 words / second)
Batch test   : 6.328274s (7901.048518 words / second)
Serial test  : 20.952494s (2386.350744 words / second)
False positive rate: 0.000000 (0.000001 expected)
Redis set add  : 0.140992s (354630.235574 words / second)
Redis pipe chk : 0.909973s (54946.676504 words / second)
Redis pipe sadd: 0.906465s (55159.324459 words / second)
Redis pipe chk : 0.887314s (56349.832669 words / second)

It is obvious that the performance result with 192.168.0.101 is much slower than 192.168.0.201.
It seems to be a bug. Any ideas?

@b4hand
Copy link
Contributor

b4hand commented Mar 7, 2016

I seriously doubt this is an issue with pyreBloom. It's much more likely an issue with your system or network configuration on the two different machines.

@Cphilo
Copy link
Author

Cphilo commented Mar 8, 2016

@b4hand , please give some test data before doubting it.

@b4hand
Copy link
Contributor

b4hand commented Mar 8, 2016

Please give some hardware and OS details, as well as the version of Redis you are using. Are there any other processes running at the same time?

Do you have any firewall rules setup on either of the machines? How are the two machines connected and what network devices are in between the two of them?

The information you've given is simply not enough information to reproduce a test.

@Cphilo
Copy link
Author

Cphilo commented Mar 8, 2016

@b4hand , add some details about my environment.
192.168.0.101 and 192.168.0.201 are on same network with 1Gb network card
redis version: Redis server v=3.0.3 redis-cli 3.0.3
os detail: Linux node1 4.2.0-16-generic #19-Ubuntu SMP Thu Oct 8 15:35:06 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux

@b4hand
Copy link
Contributor

b4hand commented Mar 8, 2016

This still doesn't tell me anything about the hardware of the two machines. For example, if you run the client and server on a single CPU machine, you would expect the performance to be worse than if you run the client and server on separate machines.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants