Pavel Zemtsov

De-multiplexing of E1 streams

• 15 Apr 2014 » De-multiplexing of E1 streams in Java
• 02 May 2014 » De-multiplexing of E1 stream: converting to C
• 12 May 2014 » The mystery of an unstable performance
• 15 May 2014 » How to make C run as fast as Java
• 18 May 2014 » Size does matter
• 04 Jun 2014 » Fast E1 de-multiplexing in C using SSE/AVX
• 21 Sep 2014 » A bug story: testing your test code
• 01 Oct 2014 » How to transpose a 16×16 byte matrix using SSE
• 21 Oct 2014 » Bug story 2: Unrolling the 16×16 matrix transposition, or be careful with macros

Performance of various callback mechanisms

• 28 Oct 2014 » Macros, call by name, lambda and Jensen's device
• 14 Nov 2014 » Lambdas and other callbacks in Java

How cache affects performance

• 24 Nov 2014 » How cache affects demultiplexing of E1
• 24 Dec 2014 » More experiments with cache

Effect of stream buffering

• 19 Jan 2015 » On the benefits of stream buffering in Java
• 01 Feb 2015 » Standard input, buffering and affinity mask

Conway's Game of Life

• 24 Apr 2015 » Game of Life, hash tables and hash codes
• 10 Sep 2015 » Hash functions: disassembling and measuring performance
• 17 Sep 2015 » Optimising division-based hash functions
• 12 Oct 2015 » The fun and the cost of refactoring
• 20 Oct 2015 » More optimisations of division-based hash: Karatsuba's method
• 28 Oct 2015 » Don't trust a micro-benchmark
• 21 Nov 2015 » CRC32: intrinsic, Java or C with JNI?
• 04 Dec 2015 » Optimising the hash-based Life
• 14 Dec 2015 » Choosing the hash map's capacity
• 31 Jan 2016 » Home-made hash tables in Java
• 21 Mar 2016 » Home-made hash tables in C++: a race against Java

Numbers from scratch

• 22 Dec 2015 » Implementing Peano arithmetic
• 10 Jan 2016 » Home-made binary numbers

Multi-threading performance

• 08 Dec 2016 » Turbo Boost and multi-threading performance
• 22 Mar 2017 » Statistical variables in Java: not quite legal but very fast

Assorted

• 05 May 2014 » One handy C idiom: multi-statement macro
• 07 May 2014 » From macros to templates
• 16 Oct 2016 » Searching in a constant set using conflict-free hashing
• 06 Nov 2016 » A bug story: data alignment on x86
• 23 Jul 2017 » The slow currentTimeMillis()
• 18 Jul 2018 » Building a fast queue between C++ and Java
• 17 Jan 2019 » To allocate or to pool?
• 13 Apr 2019 » Searching for a better indexOf
• 26 Sep 2019 » Making a char searcher in C
• 07 Oct 2019 » Abnormal string hashing

optimisation performance de-multiplexing Java C C++ macro meta-programming idiom templates GCC SSE AVX bug cache streams socket buffering stdin stdout affinity hashmap life assembly refactoring design-patterns Peano search matching alignment parallel real-time Linux Windows

• 07 Oct 2019 » Abnormal string hashing
• 26 Sep 2019 » Making a char searcher in C
• 13 Apr 2019 » Searching for a better indexOf
• 17 Jan 2019 » To allocate or to pool?
• 18 Jul 2018 » Building a fast queue between C++ and Java
• 23 Jul 2017 » The slow currentTimeMillis()
• 22 Mar 2017 » Statistical variables in Java: not quite legal but very fast
• 08 Dec 2016 » Turbo Boost and multi-threading performance
• 06 Nov 2016 » A bug story: data alignment on x86
• 16 Oct 2016 » Searching in a constant set using conflict-free hashing
• 21 Mar 2016 » Home-made hash tables in C++: a race against Java
• 31 Jan 2016 » Home-made hash tables in Java
• 10 Jan 2016 » Home-made binary numbers
• 22 Dec 2015 » Implementing Peano arithmetic
• 14 Dec 2015 » Choosing the hash map's capacity
• 04 Dec 2015 » Optimising the hash-based Life
• 21 Nov 2015 » CRC32: intrinsic, Java or C with JNI?
• 28 Oct 2015 » Don't trust a micro-benchmark
• 20 Oct 2015 » More optimisations of division-based hash: Karatsuba's method
• 12 Oct 2015 » The fun and the cost of refactoring
• 17 Sep 2015 » Optimising division-based hash functions
• 10 Sep 2015 » Hash functions: disassembling and measuring performance
• 24 Apr 2015 » Game of Life, hash tables and hash codes
• 01 Feb 2015 » Standard input, buffering and affinity mask
• 19 Jan 2015 » On the benefits of stream buffering in Java
• 24 Dec 2014 » More experiments with cache
• 24 Nov 2014 » How cache affects demultiplexing of E1
• 14 Nov 2014 » Lambdas and other callbacks in Java
• 28 Oct 2014 » Macros, call by name, lambda and Jensen's device
• 21 Oct 2014 » Bug story 2: Unrolling the 16×16 matrix transposition, or be careful with macros
• 01 Oct 2014 » How to transpose a 16×16 byte matrix using SSE
• 21 Sep 2014 » A bug story: testing your test code
• 04 Jun 2014 » Fast E1 de-multiplexing in C using SSE/AVX
• 18 May 2014 » Size does matter
• 15 May 2014 » How to make C run as fast as Java
• 12 May 2014 » The mystery of an unstable performance
• 07 May 2014 » From macros to templates
• 05 May 2014 » One handy C idiom: multi-statement macro
• 02 May 2014 » De-multiplexing of E1 stream: converting to C
• 15 Apr 2014 » De-multiplexing of E1 streams in Java