+1 (nb) from my side, I raised a few comments for CASSANDRA-15452 some time ago and Jordan addressed them. I have also backported CASSANDRA-15452 changes to my internal 4.1 fork and got about 15% reduction in compaction time even for a node with a local SSD.
On Thu, 13 Feb 2025 at 13:22, Jordan West <jw...@apache.org> wrote: > For 15452 that’s correct (and I believe also for 20092). For 15452, the > trunk and 5.0 patch are basically identical. > > Jordan > > On Thu, Feb 13, 2025 at 01:06 C. Scott Andreas <sc...@paradoxica.net> > wrote: > >> Checking to confirm the specific patches proposed for backport – is it >> the trunk commit for C-20092 and the open GitHub PR against the 5.0 branch >> for C-15452 linked below? >> >> CASSANDRA-20092: Introduce SSTableSimpleScanner for compaction (committed >> to trunk) >> https://github.com/apache/cassandra/commit/3078aea1cfc70092a185bab8ac5dc8a35627330f >> >> CASSANDRA-15452: Improve disk access patterns during compaction and >> range reads (PR available) https://github.com/apache/cassandra/pull/3606 >> >> Thanks, >> >> – Scott >> >> On Feb 12, 2025, at 9:45 PM, guo Maxwell <cclive1...@gmail.com> wrote: >> >> >> Of course, I definitely hope to see it merged into 5.0.x as soon as >> possible >> >> Jordan West <jw...@apache.org> 于2025年2月13日周四 10:48写道： >> >>> Regarding the buffer size, it is configurable. My personal take is that >>> we’ve tested this on a variety of hardware (from laptops to large instance >>> sizes) already, as well as a few different disk configs (it’s also been run >>> internally, in test, at a few places) and that it has been reviewed by four >>> committers and another contributor. Always love to see more numbers. if >>> folks want to take it for a spin on Alibaba cloud, azure, etc and determine >>> the best buffer size that’s awesome. We could document which is suggested >>> for the community. I don’t think it’s necessary to block on that however. >>> >>> Also I am of course +1 to including this in 5.0. >>> >>> Jordan >>> >>> On Wed, Feb 12, 2025 at 19:50 guo Maxwell <cclive1...@gmail.com> wrote: >>> >>>> What I understand is that there will be some differences in block >>>> storage among various cloud platforms. More intuitively, the default >>>> read-ahead size will be the same. For example, AWS ebs seems to be 256K, >>>> and Alibaba Cloud seems to be 512K（If I remember correctly). >>>> >>>> Just like 19488, give the test method, see who can assist in the test , >>>> and provide the results. >>>> >>>> Jon Haddad <j...@rustyrazorblade.com> 于2025年2月13日周四 08:30写道： >>>> >>>>> Can you elaborate why? This would be several hundred hours of work >>>>> and would cost me thousands of $$ to perform. >>>>> >>>>> Filesystems and block devices are well understood. Could you give me >>>>> an example of what you think might be different here? This is already one >>>>> of the most well tested and documented performance patches ever >>>>> contributed >>>>> to the project. >>>>> >>>>> On Wed, Feb 12, 2025 at 4:26 PM guo Maxwell <cclive1...@gmail.com> >>>>> wrote: >>>>> >>>>>> I think it should be tested on most cloud platforms（at least >>>>>> aws、azure、gcp） before merged into 5.0 . Just like CASSANDRA-19488. >>>>>> >>>>>> Paulo Motta <pa...@apache.org>于2025年2月13日 周四上午6:10写道： >>>>>> >>>>>>> I'm looking forward to these improvements, compaction needs tlc. :-) >>>>>>> A couple of questions: >>>>>>> >>>>>>> Has this been tested only on EBS, or also EC2/bare-metal/Azure/etc? >>>>>>> My >>>>>>> only concern is if this is an optimization for EBS that can be a >>>>>>> deoptimization for other environments. >>>>>>> >>>>>>> Are there reproducible scripts that anyone can run to verify the >>>>>>> improvements in their own environments ? This could help alleviate >>>>>>> any >>>>>>> concerns and gain confidence to introduce a perf. improvement in a >>>>>>> patch release. >>>>>>> >>>>>>> I have not read the ticket in detail, so apologies if this was >>>>>>> already >>>>>>> discussed there or elsewhere. >>>>>>> >>>>>>> On Wed, Feb 12, 2025 at 3:01 PM Jon Haddad <j...@rustyrazorblade.com> >>>>>>> wrote: >>>>>>> > >>>>>>> > Hey folks, >>>>>>> > >>>>>>> > Over the last 9 months Jordan and I have worked on CASSANDRA-15452 >>>>>>> [1]. The TL;DR is that we're internalizing a read ahead buffer to >>>>>>> allow us >>>>>>> to do fewer requests to disk during compaction and range reads. This >>>>>>> results in far fewer system calls (roughly 16x reduction) and on systems >>>>>>> with higher read latency, a significant improvement in compaction >>>>>>> throughput. We've tested several different EBS configurations and >>>>>>> found it >>>>>>> delivers up to a 10x improvement when read ahead is optimized to >>>>>>> minimize >>>>>>> read latency. I worked with AWS and the EBS team directly on this and >>>>>>> the >>>>>>> Best Practices for C* on EBS [2] I wrote for them. I've performance >>>>>>> tested >>>>>>> this patch extensively with hundreds of billions of operations across >>>>>>> several clusters and thousands of compactions. It has less of an >>>>>>> impact on >>>>>>> local NVMe, since the p99 latency is already 10-30x less than what you >>>>>>> see >>>>>>> on EBS (100micros vs 1-3ms), and you can do hundreds of thousands of >>>>>>> IOPS >>>>>>> vs a max of 16K. >>>>>>> > >>>>>>> > Related to this, Branimir wrote CASSANDRA-20092 [3], which >>>>>>> significantly improves compaction by avoiding reading the partition >>>>>>> index. >>>>>>> CASSANDRA-20092 has been merged to trunk already [4]. >>>>>>> > >>>>>>> > I think we should merge both of these patches into 5.0, as the >>>>>>> perf improvement should allow teams to increase density of EBS backed C* >>>>>>> clusters by 2-5x, driving cost way down. There's a lot of teams >>>>>>> running C* >>>>>>> on EBS now. I'm currently working with one that's bottlenecked on maxed >>>>>>> out EBS GP3 storage. I propose we merge both, because without >>>>>>> CASSANDRA-20092, we won't get the performance improvements in >>>>>>> CASSANDRA-15452 with BTI, only BIG format. I've tested BTI in other >>>>>>> situations and found it to be far more performant than BIG. >>>>>>> > >>>>>>> > If we were looking at a small win, I wouldn't care much, but since >>>>>>> these patches, combined with UCS, allows more teams to run C* on EBS at >>>>>>> > >>>>>>> 10TB / node, I think it's worth doing now. >>>>>>> > >>>>>>> > Thanks in advance, >>>>>>> > Jon >>>>>>> > >>>>>>> > [1] https://issues.apache.org/jira/browse/CASSANDRA-15452 >>>>>>> > [2] >>>>>>> https://aws.amazon.com/blogs/database/best-practices-for-running-apache-cassandra-with-amazon-ebs/ >>>>>>> > [3] https://issues.apache.org/jira/browse/CASSANDRA-20092 >>>>>>> > [4] >>>>>>> https://github.com/apache/cassandra/commit/3078aea1cfc70092a185bab8ac5dc8a35627330f >>>>>>> > >>>>>>> >>>>>> >> -- Dmitry Konstantinov
