ScyllaDB Blog
scylladb.com · Company blog · on HN since 2015
ScyllaDB's company blog covers its Cassandra-compatible database, the Seastar framework, benchmarks, and Linux performance topics.
What HN saysCommenters praise the team's engineering, citing Seastar's thread-per-core design and ScyllaDB's performance and scalability.
167HN people shared it
97linked it in comments
208stories submitted
3.8Kstory points
3 months agolast shared
2015people per year2026
Most-shared posts
- 1ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster27 people · 220 points · first shared 2015 · 93 comments on HN
- 2How io_uring and eBPF Will Revolutionize Programming in Linux8 people · 715 points · first shared 2020 · 320 comments on HN
- 3Project Alternator: The Scylla Open Source Amazon DynamoDB-Compatible API7 people · 35 points · first shared 2019 · 3 comments on HN
- 4The Dark Side of MongoDB’s New License7 people · 13 points · first shared 2018 · 1 comments on HN
- 5Adventures with Memory Barriers and Seastar on Linux6 people · 29 points · first shared 2018 · discussion on HN
- 6CockroachDB vs. Scylla Benchmark5 people · 31 points · first shared 2021 · 8 comments on HN
- 7Coordinated Omission (2021)5 people · 4 points · first shared 2021 · discussion on HN
- 8Scylladb: cassandra compatible NoSQL database5 people · 3 points · first shared 2015 · discussion on HN
- 9Discord Chooses Scylla as Its Core Storage Layer5 people · 3 points · first shared 2020 · discussion on HN
- 10Linux libaio async I/O is only good on XFS: Qualifying Filesystems for ScyllaDB5 people · 2 points · first shared 2016 · discussion on HN
- 11ScyllaDB – Why We're Moving to a Source Available License4 people · 96 points · first shared 2024 · 30 comments on HN
- 12Scylla's Approach to Improve Performance for CPU-Bound Workloads (2017)4 people · 77 points · first shared 2017 · discussion on HN
- 13CharybdeFS: a new fault-injecting filesystem for software testing4 people · 34 points · first shared 2016 · 6 comments on HN
- 14Complex Path for a Simple Portable Python Interpreter, or Snakes on a Data Plane4 people · 24 points · first shared 2019 · 1 comments on HN
- 15Scylla Developer Hackathon: Rust Driver4 people · 7 points · first shared 2021 · 1 comments on HN
What HN people wrote when they linked it
tilt_error · 2020-12-12
These guys are really amazing. I have been following their work for quite some time, starting with my interest in unikernels and thus OSv. Take a look at Seastar (C*) [1] and ScyllaDB [2].
Some companies are powerhouses!
[1] http://seastar.io
[2] https://www.scylladb.com
ddorian43 · 2020-07-01
Yes most software is limited cores (example: encoding videos).
The best (efficient) way to utilize that many cores IS to have 1 pinned process/thread per-core: https://www.scylladb.com/ https://github.com/scylladb/seastar/
That would be the same in Python too. A problem is that you can't share the kernel pages for the code, and you need to have a shared-cache. Probably 0-mem-copy with no deserialization example: lmdb + Flat Buffers.
Nicer is to have 1/2x cores, but each core being 2x faster ;)
Jach · 2017-10-19
Literally any other. I'm unaware of any alternative so unreliable.
But I do think http://www.scylladb.com/ is great.
erichocean · 2016-08-29
My favorite database by far today is LMDB (B+Tree).[0] Performance is insane, and very low-variance. Reads scale linearly with core counts, and it has a lot of useful index types and knobs to get maximum performance.
What am I most looking forward to using later this year? ScyllaDB[1] and CockroachDB[2], both in conjunction with LMDB.
[0] http://104.237.133.194/doc/
[1] http://www.scylladb.com/
[2] https://www.cockroachlabs.com/
PeterCorless · 2024-06-04
The best scalability I knew about was JanusGraph running on ScyllaDB. You can also run JanusGraph on Cassandra, which is horizontally scalable. ScyllaDB will be more performant and vertically scalable.
• Docs: https://docs.janusgraph.org/storage-backend/scylladb/
• Detailed use case: https://www.scylladb.com/2019/05/14/powering-a-graph-data-system-with-…
• Video: https://www.youtube.com/watch?v=7WZyVUTwYJ4
Note this was a few years ago. I haven't seen a TigerGraph vs. JanusGraph head-to-head. The TigerGraph people are pretty sharp. If anyone has deeper knowledge would love to see comparative benchmarks.
Both JanusGraph and TigerGraph are rated about the same in terms of popularity on DB-engines.com (between 100-150 rankings). Neo4j is still far more popular/well-known, at rank #21 this past month. But there are other options you can explore.
humzashahid98 · 2024-05-22
B+ Trees (which store data at leaves) rather than B-Trees are what you want for good iteration performance usually.
An ordered B-Tree iteration would jump between different arrays in the tree often, but a B+ Tree would only have you jump to another array once you’re done iterating over all the elements in the current array.
The following article says the same. https://www.scylladb.com/2021/11/23/the-taming-of-the-b-trees/#:~:text…
ofrzeta · 2022-09-30
There's ScyllaDB as a replacement for Cassandra. https://www.scylladb.com/
DenseComet · 2022-03-29
Take a look at ScyllaDB Alternator [1], which is API compatible with DynamoDB. Scylla also supports multiregion clusters, which should help with latency. I've not used alternator, but I've had a good experience with Scylla, and it might be worth looking into as a lower effort way to move off DynamoDB.
[1] https://www.scylladb.com/alternator/
akmittal · 2022-03-23
Last I heard was from https://www.scylladb.com/
Selling point being C++ version of cassandra
xemul · 2022-02-18
I wouldn't say RB-tree rebalancing is much simpler. On the other hand, in B-trees there's a whole lot of poorly documented nuances, binary trees are more intuitive. But in the end of the day the implementation effort perfectly worth it. https://www.scylladb.com/2021/11/23/the-taming-of-the-b-trees/
thekozmo · 2022-01-20
Indeed quite a journey. If you love DynamoDB and like open source, give Scylla a try (disclosure me==founder): https://www.scylladb.com/alternator/
staticassertion · 2021-09-27
https://www.scylladb.com/2021/08/24/apache-cassandra-4-0-vs-scylla-4-4…
I suspect this talk is a follow up to this blog post, which may be preferable for many.
traceroute66 · 2021-09-12
> Cloud-managed NoSQL databases is the potential for zero-downtime maintenance.
Aah yes... that magical "zero-downtime" cloud, lots of promises are made by the cloud providers in that area but all eventually broken at some point. Its a fact of life with technology unfortunately. :)
No need for the cloud with ScyllaDB[1], CockroachDB[2] and others all being used in production.
LetsEncrypt chose their setup for perfectly valid reasons they described on their blog[3]. Please describe your expertise for saying they're "doing it wrong", because its clear from their blog they put a lot of time thinking about it.
[1]https://www.scylladb.com/
[2]https://www.cockroachlabs.com/
[3]https://letsencrypt.org/2021/01/21/next-gen-database-servers.html
ctvo · 2021-08-30
Scylla [1] is a cloud key / value store competing directly with DynamoDB. They have an interface identical to DDB to make transitioning easier [2]. They outperform DDB for many tasks, and have many added features.
My point is this often is AWS, Azure, GCP vs. start-up, but there are counter examples.
1 - https://www.scylladb.com
2 - https://www.scylladb.com/alternator
englambert · 2021-02-01
Chris here, from the Opstrace team. As it turns out, it’s just a happy coincidence. When we discovered theirs we fell in love with it as well. They have many different versions of their monster (https://www.scylladb.com/media-kit/)... similarly you’ll see several new versions of our mascot, Tracy the Octopus, over time!
arianvanp · 2021-02-01
Your mascot is almost exactly identical to https://scylladb.com/ 's mascot. Is there any connection; or a happy accident?
StreamBright · 2020-05-17
>> Easier for who?
For any decent SRE out there.
>> magnitudes more performance which is not possible by just tuning Cassandra
Magnitudes?? Are you talking about the order of magnitude? You should read the ScyllaDB performance report first.
https://www.scylladb.com/product/benchmarks/aws-i3-metal-benchmark/
Avg. 99.9% Latency (ms): 9.9
vs.
Avg. 99.9% Latency (ms): 474.4
While there is no significant latency difference in lover percentile tiers. Where Scylla really shines is TCO. Some companies trade SRE time for license cost, some other companies tune GC.
eyberg · 2020-05-16
There is plenty of work in this area. The team behind OSv ending up writing https://www.scylladb.com/ because of their earlier interest in databases.
adev_ · 2020-02-05
> The JVM world tends to solve this problem by using off-heap caches. See Apache Ignite [0] or Ehcache [1].
Yeah, but I really do not bite your argument.
When you are reduced to do manual memory management and fight the GC of your language, maybe you should simply not use a language with GC in the first place.
They are right to use rust ( or C/C++) for that. It's not for nothing that redis (C) is so successful in the LRU domain.
> It's worth mentioning that Apache Cassandra itself uses an off-heap cache.
And still ScyllaDB (C++) is able to completely destroy Cassandra in term of AVG latency [0]
[0]: https://www.scylladb.com/product/benchmarks/
wenc · 2019-11-01
Scylla [1] for instance is a C++ rewrite and a drop-in replacement of JVM-based Cassandra, and from what I've read is fairly stable and performs much faster.
[1] https://www.scylladb.com/
ssvss · 2019-10-07
Most DBAs working with async/directIO databases recommend XFS,
https://www.scylladb.com/2016/02/09/qualifying-filesystems/
mlla · 2018-03-06
Good Q. You might like to check ScyllaDB written in C++, which is supposed to have considerably better performance than Cassandra (also low tail-latency) and a level of compatibility with it: https://www.scylladb.com/
ehllo · 2017-12-01
Maybe this is a better option for you.
http://www.scylladb.com/
manigandham · 2017-05-16
Also take a look at https://www.pingcap.com/ for sql scale-out and http://www.scylladb.com/ for cassandra-but-fast.
dragonne · 2017-04-22
Scylla's blog has some excellent technical content, though it's intermixed with lots release announcements. http://www.scylladb.com/2017/01/02/top-5-blog-posts-of-2016/ has some of the highlights.
sandGorgon · 2017-02-01
I used to think this too, until I came across http://www.scylladb.com/
It is a fork of cassandra written in the Seastar c++ framework and is drop-in compatible with cassandra. Claims 10x increase in performance.
I always thought there was a few percentage points difference - never a 10x performance difference between java and c++. And that too for a project with as many man hours and facebook-scale tuning as cassandra.
espeed · 2017-01-23
See LMDB (embedded) and ScyllaDB (distributed):
The Lightning Memory-Mapped Database (LMDB) https://www.youtube.com/watch?v=Rx1-in-a1Xc
ScyllaDB http://www.scylladb.com/
whitegrape · 2016-03-31
Has anyone done a meaningful private benchmark comparison with http://www.scylladb.com/ ? I didn't find one online.
pella · 2015-10-14
or with http://www.scylladb.com/ backend. "ScyllaDB: world's fastest NoSQL column store database; Fully compatible with Apache Cassandra at 10x the throughput and jaw dropping low latency"
nfa_backward · 2015-09-29
From my experience and the experience of others ( https://www.eecs.berkeley.edu/~keo/publications/nsdi15-final147.pdf ) current big data solutions are more often CPU bound not IO. I think that we will be seeing more and more of big data architecture moving to C++. For example: http://www.scylladb.com/