Top all-time · #7299
Mikael Ronstrom
Personal blog · mikaelronstrom.blogspot.comdatasystems
Mikael Ronstrom's personal blog at mikaelronstrom.blogspot.com covers data and systems. Since 2008, 11 different Hacker News readers have linked it 31 times, most often “DB thread pipelines reduce context switches, instructions/cycle, cache misses”. It is still writing: the last post was on 2026-09-22, and you can follow it by feed.
Too few sampled comments (8) to say how often HN recommends it.?
Why HN links it
HN readers link this blog for benchmark results and technical notes on MySQL Cluster and RonDB (NDB Cluster), including thread design, YCSB and sysbench tests across AWS, GCP and Azure.
Summarised by Claude from a sample of 8 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
My name is Mikael Ronstrom and I work for Hopsworks AB as Head of Data. I also assist companies working with NDB Cluster as self-employed consultant. I am a member of The Church of Jesus Christ of Latter Day Saints. I can be contacted at mikael dot ronstrom at gmail dot com for NDB consultancy servi
Most-linked posts?
- DB thread pipelines reduce context switches, instructions/cycle, cache misses2 people · 3 mentions · first linked 2021-03-16
- The Desktop is back (for database development, at least)2 people · 2 mentions · first linked 2023-03-23
- Research on Thread Pipelines Using RonDB2 people · 2 mentions · first linked 2021-05-10
What Hacker News says?
Some of the stuff in here is just wrong. "they have to block --- wait until the coordinator recovers --- in order to find out the final decision" This assumes there is only one coordinator in the system and that there cannot be another coordinator that 'takes over'. Here's a good example of a real-world 2PC system that is non-blocking if the coordinator fails - NDB: http://mikaelronstrom.blogspot.com/2018/09/non-blo…
Comment linking /2018/09/non-blocking-two-phase-commit-in-ndb.html
on “It’s Time to Move on from Two Phase Commit”Submitted How to design a DBMS for Telco requirements as a story · 33 points
HN discussionYou are correct that they serialize cross-partition transactions. So if your workload has a reasonable percentage of write/update/delete operations it is possible that you will bottleneck on the global coordinator. From this blog https://fauna.com/blog/distributed-acid-transaction-performa... , you can get 3300 transactions/second. Daniel Abadi claims you can get to 500,000 trans/sec - http://dbmsmusings.blogspot.co…
Comment linking /2015/03/200m-reads-per-second-in-mysql-cluster.html
on “Consistency Without Clocks: FaunaDB's Distributed Transaction Protocol”Submitted Automatic memory management in RonDB as a story · 14 points
HN discussionWhy should I use this over MySQL Cluster? MySQL Cluster is lock-free, distributed, HA, and has better performances: http://mikaelronstrom.blogspot.com/2012/05/mysql-cluster-72-... http://mikaelronstrom.blogspot.com/2012/05/mysql-cluster-727... And MySQL Cluster has been used in mission critical apps for years. What does MemSQL give me that MySQL Cluster doesn't?
Comment linking /2012/05/mysql-cluster-727-achieves-1bn-update.html
on “Ex-Facebookers launch MemSQL (YC W11) to make your database fly”Submitted DB thread pipelines reduce context switches, instructions/cycle, cache misses as a story · 10 points
HN discussionRonDB (NDB Cluster) takes a different approach to threading and claims it's faster than scylladb-style sharding http://mikaelronstrom.blogspot.com/2021/03/designing-thread-...
Comment linking /2021/03/designing-thread-pipeline-for-optimal.html
on “KeyDB CEO Interview: Getting into YC with a Fork of Redis”Submitted Perf Benchmark AWS/GCP/Azure with Sysbench OLTP and RonDB as a story · 6 points
HN discussionAs a MySQL, Oracle and Clustrix DBA this is very sexy news. Here is the extra info on the benchmark http://mikaelronstrom.blogspot.com/2012/02/105bn-qpm-using-m... But no info on what disk setup they used. But for every instance node there was a storage node.
Comment linking /2012/02/105bn-qpm-using-mysql-cluster-72.html
on “MySQL Cluster 7.2 GA Released, Delivers 1 Billion Queries per Minute”Submitted Benchmarking a 5 TB Data Node in NDB Cluster as a story · 6 points
HN discussionYCSB is one benchmark that is commonly used by key-value stores. See http://mikaelronstrom.blogspot.com/2020/10/ycsb-disk-data-be... , http://mikaelronstrom.blogspot.com/2020/02/ndb-cluster-world...
Comment linking /2020/02/ndb-cluster-worlds-fastest-key-value.html
on “Show HN: RonDB – fast key-value database in the cloud”Submitted NDB Cluster, the World's Fastest Key-Value Store as a story · 5 points
HN discussionAssuming sacrificing: The availability numbers requires instant failover, this requires updating all replicas synchronously. Throughput and latency are both coming from using an asynchronous programming model which have been refined over the years. Todays new blog on this topic is here: http://mikaelronstrom.blogspot.com/2021/05/research-on-threa...
Comment linking /2021/05/research-on-thread-pipelines-using-rondb.html
on “Show HN: RonDB – fast key-value database in the cloud”Submitted Ha vs. AlwaysOn as a story · 4 points
HN discussionSome sysbench benchmark results are here comparing results on AWS, GCP, and Azure. Interesting to see AWS is better at lower low load (due to better interrupt handling in VMs?), while GCP is best at highest loads. http://mikaelronstrom.blogspot.com/
Comment linking /
on “Show HN: RonDB – fast key-value database in the cloud”Submitted Introducing MySQL Cluster 7.6 as a story · 3 points
HN discussionRonDB is open-source and supports on-disk data on NVMe disks. http://mikaelronstrom.blogspot.com/2022/04/variable-sized-di...
Comment linking /2022/04/variable-sized-disk-rows-in-rondb.html
on “Ask HN: Why are there no open source NVMe-native key value stores in 2023?”Submitted RonDB receives ARM64 support and large transaction support as a story · 2 points
HN discussionCheck out NDB Cluster (Rondb) and the blog of the creator, example of a thread pipeline to be more efficient than Redis/ScyllaDB per-core sharding: http://mikaelronstrom.blogspot.com/2021/03/designing-thread-...
Comment linking /2021/03/designing-thread-pipeline-for-optimal.html
on “Ask HN: What is the current state of art in distributed databases/datastores?”Submitted Scaling HDFS with MySQL Cluster for Metadata as a story · 2 points
HN discussionnot a lot to go on, start here http://hackmysql.com/documents http://dev.mysql.com/doc/refman/5.0/en/optimization.html http://mysqldba.blogspot.com/2008/04/common-steps-to-scale-l... http://mikaelronstrom.blogspot.com/2008/07/1-making-mysql-cl...
Comment linking /2008/07/1-making-mysql-cluster-scale-perfectly.html
on “Ask HN: A few hundred users and MySQL started to complain...”Submitted We used coding agents to add RonSQL support to RonDB as a story · 1 points
HN discussionSubmitted How to reach 100M Key lookups using REST server with Python clients as a story · 1 points
HN discussionSubmitted The completion of a 12 year long project in RonDB as a story · 1 points
HN discussionSubmitted The Desktop is back (for database development, at least) as a story · 1 points
HN discussionSubmitted Laptop vs. Desktop for RonDB Development as a story · 1 points
HN discussionSubmitted Research on Thread Pipelines Using RonDB as a story · 1 points
HN discussionSubmitted Utilizing better data and instruction cache on RonDB threading model as a story · 1 points
HN discussionSubmitted How to make an efficient Scalable Key Lookup engine of MySQL Cluster as a story · 1 points
HN discussion
