zX41ZdbW

53 ms·

631.
by zX41ZdbW 4y ago
The list of references is incomplete without: https://clickhouse.com/blog/testing-the-performance-of-click...
632.
by zX41ZdbW 4y ago
It is nice to see more on "build your own database" topics. Although people constantly bug me on attempts to build their own ClickHouse. Someone is trying to do it with Apache Arrow and DataFusion. Folks from DuckDB are trying to
633.
by zX41ZdbW 4y ago
I found Citus implementation quite weak among analytic databases. It does not fare well even among Postgres-based systems, like Greenplum [1]. [1] https://benchmark.clickhouse.com/
634.
by zX41ZdbW 4y ago
Redshift is an OLAP database, much like ClickHouse. For OLAP databases, horizontal scalability is a must - due to generally larger data volumes. While Postgres is OLTP. For OLTP databases achieving horizontal scalability require a more soph
635.
by zX41ZdbW 4y ago
Looking forward to seeing a post describing the integration of ClickHouse with MinIO and cost/performance comparison between the usage of AWS S3 and MinIO. https://clickhouse.com/docs/en/integrations/s3&#
636.
by zX41ZdbW 4y ago
Although it has its gotchas, MySQL dialect is still somewhat better from the usability standpoint. "SHOW TABLES" is nice; it is natural and self-explanatory. "\d" is obscure. It also reminds me how many DBMSs did not sup
637.
by zX41ZdbW 4y ago
Simply by larger sizes of compressed blocks, which are limited to page size in Postgres, and by improving the data locality by sorting, which is inherent for LSM-trees. But if you want higher compression, you need to consider column-oriente
638.
by zX41ZdbW 4y ago
If Postgres were ClickHouse, it would refresh materialized views synchronously on INSERT... but for Postgres it isn't.
639.
by zX41ZdbW 4y ago
I'd like to emphasize the importance of built-in protection from human and configuration errors. For example, if you are trying to drop a table of large enough size, ClickHouse will ask to defuse the protection first: https:/&#x2
640.
by zX41ZdbW 4y ago
I also found it strange that PostgreSQL cannot simply support data formats of previous versions. In contrast, the latest version of ClickHouse[1] (23.1) can be installed over the version from 2016, and it does not require format conversions
641.
by zX41ZdbW 4y ago
Worth to note that there is no innovation in DuckDB. The features such as: SELECT * EXCLUDE SELECT * REPLACE Column Aliases in WHERE / GROUP BY / HAVING Struct Dot Notation Function Aliases from Other Databas
642.
by zX41ZdbW 4y ago
What is the purpose of "Implementing Database Systems with Apache Arrow and DataFusion" instead of using ClickHouse, which does all the stuff?
643.
by zX41ZdbW 4y ago
Syscalls can be heavier than expected. One example is when an application is run inside gVisor. Another example is when a lot of eBPF code is attached. A third example is when a program is run under strace. Disclaimer: I'm working on C
644.
by zX41ZdbW 4y ago
ClickHouse has the `transform` function for this purpose: SELECT transform(id, [1, 2, 11], ['ClickHouse', 'Postgres', 'MongoDB'], '') It creates a lookup table. When you are using the CASE op
645.
by zX41ZdbW 4y ago
We are also interested in the comparison of Hudi, Delta Lake, and Iceberg. We have recently added support for Hudi and Delta Lake; you can check here: https://clickhouse.com/docs/en/engines/table-engines/
646.
by zX41ZdbW 4y ago
There is no such thing as the fastest sorting algorithm. Sorting is a landscape of hundreds of algorithms, with various details and assumptions. Even the definition of the sorting task can be different - depending on what, how, and why you
647.
by zX41ZdbW 4y ago
It is a nice paper, but for practical applications on fixed-size numbers up to 64 bits, as well as for tuples on fixed-size numbers, and for partial sorting, it does not beat Radix Sort (on moderate-sized arrays). I have tested it in ClickH
648.
by zX41ZdbW 4y ago
The relations between Hilbert Curves[1], Gray Code, De Bruijn sequences[2]... https://en.wikipedia.org/wiki/Hilbert_curve https://en.wikipedia.org/wiki/De_Bruijn_sequence I want to make use of it
649.
by zX41ZdbW 4y ago
These formats look like an attempt to get a halfway solution: you want to get something like a real MPP analytic DBMS (e.g., ClickHouse) but have to use a data lake for some reason. It resembles previous trendy technologies that are mostly
650.
by zX41ZdbW 4y ago
I'm successfully* using it in a few personal projects: ClickHouse Playground: https://play.clickhouse.com/play?user=play#U0VMRUNUIDE= ClickHouse Dashboards: https://sql.clickhouse.com/dashboard#eyJob3N0
651.
by zX41ZdbW 4y ago
Memory bound alternative for the COUNT(DISTINCT ...) aggregate function in SQL. ClickHouse[1] has a few functions for that, with different tradeoffs in memory usage, precision, and performance: - uniqExact - the same as COUNT(DISTINCT
652.
by zX41ZdbW 4y ago
Content Defined Chunking is one of my favorite algorithms because it has some "magic" similar to HyperLogLogs, Bloom filters, etc... This algorithm is good to explain to people, to get them inspired by computer science. I usually
653.
by zX41ZdbW 4y ago
If you downloaded `clickhouse-local`, you can simply rename it to `clickhouse-server` and run. It will automatically transform itself to ClickHouse server.
654.
by zX41ZdbW 4y ago
I did say about > Query language, data types support, feature completeness, stability and testing nothing about correctness. In terms of stability, I see a couple of pretty old, and still unresolved issues about memory safety (data races
655.
by zX41ZdbW 4y ago
Totally agree. But it's usually more complicated. For example, MongoDB supports Ubuntu 20.04, but not 22.04: https://www.mongodb.com/docs/manual/tutorial/install-mongodb... Or another example - folks in
656.
by zX41ZdbW 4y ago
ClickHouse will throw an exception in case of not enough memory and continue to serve other queries. Under certain configurations it can OOM as well. > How confident are you that your chosen dataset is neutral? I have no idea if it is &q
657.
by zX41ZdbW 4y ago
This is not what we want. I admit this problem. Changes will be made. Just for amusement, here is the 2016 website version: https://presentations.clickhouse.com/original_website/
658.
by zX41ZdbW 4y ago
+ totally agree with that. There is a use case when you can improve performance by keeping compressed (LZ4) data in RAM and decompressing by small blocks that fit in cache. This is demonstrated by ClickHouse[1][2] - the whole data processin
659.
by zX41ZdbW 4y ago
Recently I tried to load another dataset to DuckDB: https://github.com/ClickHouse/ClickHouse/issues/42363 for my own entertainment. But it did not work (the server became unresponsive after consuming all memo
660.
by zX41ZdbW 4y ago
This summer I was preparing the ClickBench: https://benchmark.clickhouse.com/ When I tried to use DuckDB on the same dataset as ClickHouse, it simply did not work due to OOM: https://github.com/duckdb/d

More ›