HNHacker News
TopNewBestAskShowJobs

mikaelronstrom

34 karma · joined April 9, 2018

submissionscomments
mikaelronstrom··on Show HN: RonDB – fast key-value database in the cloud
The release notes of RonDB 21.04.0 found in the RonDB documentation at docs.rondb.com lists all the differences compared to NDB.
mikaelronstrom··on Perf Benchmark AWS/GCP/Azure with Sysbench OLTP and RonDB
On latency AWS > GCP, Azure

On throughput GCP > AWS, Azure

mikaelronstrom··on RonDB: The fastest key-value store, made for AI/ML- now in the cloud
Another metaphorical way of describing the difference is that NDB is high-end PC whereas RonDB is a powerful Mac. NDB can be configured in any number of ways whereas RonDB does the configuration automatically and the user will always get access to the most advanced features of NDB.
mikaelronstrom··on RonDB: The fastest key-value store, made for AI/ML- now in the cloud
Also an additional feature in ClusterJ that improves key-value applications using batched read by more than 3x.
mikaelronstrom··on RonDB: The fastest key-value store, made for AI/ML- now in the cloud
As mentioned there will be more blogs. But we focus on the LATS (low Latency, high Availability, high Throughput and scalable Storage). Other key-value stores have some of the LATS qualities but RonDB has them all. SQL capabilities on top of this helps.
mikaelronstrom··on RonDB: The fastest key-value store, made for AI/ML- now in the cloud
NDB Cluster 8.0.23 contains a new thread model that makes it possible to grow and shrink size of data nodes. This is default in RonDB, optional in NDB. Also RonDB has fully automated the memory configuration such that the user needs only specify which VM instance type he wants and the number of VMs. Plus a number of bug fixes, improvement to monitoring and some more.
mikaelronstrom··on Nvidia to Acquire Mellanox for $6.9B
Mellanox is big in high-end Ethernet equipment as well which is growing faster than the Infiniband business, obviously a key technology for cloud providers.
mikaelronstrom··on CockroachDB 2.0 Performance Makes Significant Strides
Open source DBMS normally provide numbers using some open source variant of TPC-C. One such implementation is DBT2. TPC-C according to the standard is extremely expensive to setup and is only interesting for databases with massive disks. However most open source DBMS runs DBT2 with no wait time, this means that also an in-memory DBMS can report numbers on a TPC-C-like benchmark (e.g. DBT2).

MySQL Cluster (NDB) did such benchmarks a few years ago where we ran 2.2M transactions per minute using DBT2. This meant executing millions of SQL queries per second towards the cluster and most of the CPU load is in the set of MySQL servers.

Currently I am using the load phase of this benchmark to test loading massive amounts of data into the new 7.6 version of MySQL Cluster using machines with 1 TB of memory.

← PreviousPage 2 of 2