starred/s3fs-fuse

Fork 0

mirror of https://github.com/s3fs-fuse/s3fs-fuse.git synced 2026-04-25 13:26:00 +03:00

[GH-ISSUE #717] S3FS Getting A Performance Boost From Elastic Network Adapter (ENA) ? #406

New issue

Closed

opened 2026-03-04 01:45:15 +03:00 by kerem · 6 comments

kerem commented

2026-03-04 01:45:15 +03:00

Owner

Originally created by @tspicer on GitHub (Feb 7, 2018).
Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/717

Looks like if setup correctly, the pipe between EC2 and S3 just got a big upgrade. I was curious how this impacts S3FS. My assumption is that it would realize the benefits of this upgrade, but maybe it is not so simple :)

"Traffic to and from Amazon Simple Storage Service (S3) can now take advantage of up to 25 Gbps of bandwidth. Previously, traffic of this type had access to 5 Gbps of bandwidth. This will be of benefit to applications that access large amounts of data in S3 or that make use of S3 for backup and restore."

https://aws.amazon.com/blogs/aws/the-floodgates-are-open-increased-network-bandwidth-for-ec2-instances/

Originally created by @tspicer on GitHub (Feb 7, 2018). Original GitHub issue: https://github.com/s3fs-fuse/s3fs-fuse/issues/717 Looks like if setup correctly, the pipe between EC2 and S3 just got a big upgrade. I was curious how this impacts S3FS. My assumption is that it would realize the benefits of this upgrade, but maybe it is not so simple :) "Traffic to and from Amazon Simple Storage Service (S3) can now take advantage of up to 25 Gbps of bandwidth. Previously, traffic of this type had access to 5 Gbps of bandwidth. This will be of benefit to applications that access large amounts of data in S3 or that make use of S3 for backup and restore." https://aws.amazon.com/blogs/aws/the-floodgates-are-open-increased-network-bandwidth-for-ec2-instances/

kerem closed this issue

2026-03-04 01:45:15 +03:00

kerem commented

2026-03-04 01:45:16 +03:00

Author

Owner

@gaul commented on GitHub (Feb 7, 2018):

That is a lot of bandwidth! I suspect that s3fs needs some tuning to take advantage of this. @orozery added some performance improvements a few months ago so I encourage you to test with the latst version and report back. It would also be interesting to compare with goofys.

@gaul commented on GitHub (Feb 7, 2018): That is a lot of bandwidth! I suspect that s3fs needs some tuning to take advantage of this. @orozery added some performance improvements a few months ago so I encourage you to test with the latst version and report back. It would also be interesting to compare with [goofys](https://github.com/kahing/goofys).

kerem commented

2026-03-04 01:45:16 +03:00

Author

Owner

@tspicer commented on GitHub (Feb 7, 2018):

Was not familiar with goofys so thanks for the tip. In terms of testing i will look at what would be involved and report back once I have some time to undertake it.

@tspicer commented on GitHub (Feb 7, 2018): Was not familiar with goofys so thanks for the tip. In terms of testing i will look at what would be involved and report back once I have some time to undertake it.

kerem commented

2026-03-04 01:45:16 +03:00

Author

Owner

@MaartenW commented on GitHub (Jun 14, 2018):

Sounds very interesting. Were you able to allocate some time to investigate?

@MaartenW commented on GitHub (Jun 14, 2018): Sounds very interesting. Were you able to allocate some time to investigate?

kerem commented

2026-03-04 01:45:16 +03:00

Author

Owner

@tspicer commented on GitHub (Jun 15, 2018):

I have not. We have prioritized stability vs speed at the moment so I have not looked at this

@tspicer commented on GitHub (Jun 15, 2018): I have not. We have prioritized stability vs speed at the moment so I have not looked at this

kerem commented

2026-03-04 01:45:16 +03:00

Author

Owner

@gaul commented on GitHub (Jan 26, 2019):

You should test with master which has some performance improvements. Also see #743 which suggests tuning -o multipart_size.

@gaul commented on GitHub (Jan 26, 2019): You should test with master which has some performance improvements. Also see #743 which suggests tuning `-o multipart_size`.

kerem commented

2026-03-04 01:45:16 +03:00

Author

Owner

@gaul commented on GitHub (Jun 25, 2019):

Newer versions of s3fs improve write speed. #941 tracks automatically tuning the part sizes. We do expect AWS ENA to improve performance.

@gaul commented on GitHub (Jun 25, 2019): Newer versions of s3fs improve write speed. #941 tracks automatically tuning the part sizes. We do expect AWS ENA to improve performance.