Presenting LiteSpeed’s Fast Huffman Encoder, our optimization tricks, and some benchmarks

One reason LiteSpeed outperforms nginx in HTTP/2 benchmarks is that LiteSpeed’s HPACK Huffman decoder is twice as fast as that used by nginx. Learn how our Huffman decoder works.

Explicit Congestion Notification The second day of the Kista Interim Meeting …

The QUIC Working Group is trying to make the November deadline, looking to reach consensus on as many issues as possible.

OpenLiteSpeed users should be familiar with the CGI executable lsphp. Now there is a module available: mod_lsphp. Learn about the pros and cons of switching to a module.

The Melbourne Interim Meeting has come and gone. The schedule was packed tight with topics as the Working Group strives to hit the November 2018 deadline for QUIC v1.

QCRAM-04 addresses the deadlock hazard, but is unable to minimize HoLB for a certain class of applications. Learn more.

One way to prevent attack is to not read from the stream until the decoder can process the stream immediately. This approach creates a deadlock hazard and is examined in this article.

Compression schemes that allow head-of-line blocking are targets for memory exhaustion attacks against the decoder. One way to avoid an attack is presented.