Open-source distributed SQL Database
Joined February 2022
- Tweets61
- Following0
- Followers281
- Likes22
YDBPlatform retweeted
io_uring easily beats AIO and gets faster with every kernel — until both suddenly get 30% slower.
Join a database developer’s unexpected journey into the Linux kernel and IOMMU.
medium.com/ydbtech/how-io-ur…
YDBPlatform retweeted
That's a very well-written paper from 2020 focusing on decoupling storage and compute in a distributed database. In their approach, they are very close to YDB. In YDB, the "Database layer" is built on top of distributed storage, which is somewhat similar to Hailstorm, because YDB fully disaggregates compute and storage - the main goal of Hailstorm. I would like to highlight some key differences between Hailstorm and YDB ⬇️
Turns out people *are* in fact still thinking about distributed filesystems.
eecg.toronto.edu/~ashvin/pub…
YDBPlatform retweeted
A very nice and informative post on secondary indexes in distributed databases. The only drawback is missing @YDBPlatform. YDB has an interesting approach with indexes: it offers a choice to its applications supporting both synchronous secondary indexes (similar to YugabyteDB, CockroachDB and TiDB, all mentioned in the post) as well as asynchronous secondary indexes. The consistency / performance tradeoff between these two options is described in the post in depth.
✍️ Wrote a new database post!
This one is about the different ways that distributed databases handle secondary indexes.
@FranckPachot got me thinking about this topic a few weeks back after I wrote a piece on DynamoDB secondary indexes for @RocksetCloud.
Basically, distributed databases want to shard your data onto different machines. They use a shard key / function to determine which machine holds a given record.
But what happens when you have secondary indexes and queries that don't use your shard key?
Basically use the following flowchart:
First, does the database reshard items into new shards for secondary indexes?
- If yes, is resharding done synchronously during writes?
- If yes, then you have Yugabyte, Spanner, TiDB, etc.
- If no, then you have DynamoDB GSIs or Rockset
- If it's not resharded, are queries allow to span multiple shards?
- If yes, then you have MongoDB, Vitess, Cassandra, Elasticsearch
- If no, then you have DynamoDB LSIs
Walked through the benefits and drawbacks of each as well.
Let me know where I'm wrong!
YDBPlatform retweeted
I'm thrilled to share with you an article "When Postgres is not enough: performance evaluation of PostgreSQL vs. Distributed DBMSs". Dive into an empirical study of @PostgreSQL, @CockroachDB and @YDBPlatform with the TPC-C benchmark and explore the intriguing trade-offs between reliability and performance in Postgres.
blog.ydb.tech/when-postgres-…
YDBPlatform retweeted
In my latest post, I share my experience with #Java 21 virtual threads. I successfully transitioned the TPC-C benchmark from @CMUDB to virtual threads and encountered practical examples of deadlocks. Additionally, I shortly discuss fundamental aspects of sync/async programming and highlight the benefits of virtual threads.
Explore the details in the post: bit.ly/java-21-virtual-threa…
YDBPlatform retweeted
youtu.be/iMjq96GQTHQ
In this talk, I discuss YDB performance, compare it to CockroachDB and YugabyteDB, explain the importance of choosing distributed database for mission-critical data as opposed to single-server database.
#YDB #CockroachDB #YugabyteDB #YCSB
YDBPlatform retweeted
I'm speaking at Data TLV Summit September, 15, 2022. @oleshiy and me are presenting YDB @YDBPlatform #datatlvsummit
YDBPlatform retweeted
Last week's presentation and discussion on YandexDB (YDB) is up on YouTube! Thanks, @fomichev for presenting!
youtube.com/watch?v=eYpyKXih…
