Dijkstra's handwritten presentation of the Fast Fourier Transform https://www.cs.utexas.edu/~EWD/ewd08xx/EWD807.PDF

2 replies

0 recast

9 reactions

https://bford.info/pub/net/p2pnat-abs/

1 reply

0 recast

9 reactions

Implementing complex numbers and FFT with just datatypes (no floats) https://gist.github.com/VictorTaelin/5776ede998d0039ad1cc9b12fd96811c absolutely goated

1 reply

0 recast

7 reactions

https://chipsandcheese.com/p/zen-5s-avx-512-frequency-behavior Looks like AMD is increasingly the correct choice for vectorized workloads.

0 reply

0 recast

3 reactions

Neat site on learning FPGAs: https://nandland.com

0 reply

0 recast

3 reactions

https://stackoverflow.com/a/8391601 Fascinating code to get max utilization from a cpu

1 reply

0 recast

1 reaction

https://www.phoronix.com/review/amd-epyc-9965-9755-benchmarks/3 Interesting to see the comparison of the 9965 (more cores) and the 9755 (higher freq.) on ClickHouse and RocksDB tests. More cores helps random reads while higher freq. presumably helped all of CH's compression and vectorized query processing. I wonder if the CPUs have different io characteristics.

2 replies

0 recast

3 reactions

A while back I wondered if SSD vendors would start building for specific database implementation. Apparently the gods are on our side https://blocksandfiles.com/2024/10/16/kioxia-speeds-rocksdb-1-8x-with-flexible-data-placement-ssd/ https://warpcast.com/ryansmith/0x40563ad3

0 reply

1 recast

4 reactions

Went on a little tour of IPMI/BMC security discussions. A few notes - Seems more problematic at scale when you have many servers and require remote management - You don't need to worry about network security if no one can access the network (ie don't plug a cable into the management port) Schneier has a good summary post: https://www.schneier.com/blog/archives/2013/01/the_eavesdroppi.html

0 reply

0 recast

3 reactions

Have any of you had experience running SMCI? If so would you do it again?

3 replies

2 recasts

13 reactions

TIL Prolog is extremely well suited for developing web applications.

1 reply

3 recasts

5 reactions

“People who are really serious about software should make their own hardware.”

3 replies

2 recasts

9 reactions

https://transactional.blog/blog/2024-modern-database-hardware Very excited for the future. I wonder how many new DB features will be implemented within the SSD.

0 reply

2 recasts

4 reactions

Years ago I looked at HAMMER as an alt for ZFS. At the time it lacked features like compression. It looks like the default for Dragonfly BSD is now HAMMER2 -- which now has compression, dedup, etc. tldr DragonFlyBSD is very interesting: https://www.dragonflybsd.org/history/

0 reply

3 recasts

1 reaction

Fun thought snack from SICP: Allowing quotation in a language wreaks havoc with the ability to reason about the language in simple terms, because it destroys the notion that equals can be substituted for equals. For example, three is one plus two, but the word “three” is not the phrase “one plus two.” Quotation is powerful because it gives us a way to build expressions that manipulate other expressions (as we will see when we write an interpreter in [[Chapter 4]]). But allowing statements in a language that talk about other statements in that language makes it very difficult to maintain any coherent principle of what “equals can be substituted for equals” should mean. For example, if we know that the evening star is the morning star, then from the statement “the evening star is Venus” we can deduce “the morning star is Venus.” However, given that “John knows that the evening star is Venus” we cannot infer that “John knows that the morning star is Venus.”

0 reply

7 recasts

11 reactions