More »

Efficient low-level VMs implemented in high-level languages

8 months ago by Mauricio Fernandez on Eigenclass.
The 2006 edition of the ICFP programming contest, one of the most enjoyable to date, introduced the Universal Machine used by a fictional society to program...

Hash tables: separate chaining vs. double hashing

8 months ago by Mauricio Fernandez on Eigenclass.
This entry uses jsMath to display mathematical expressions. Please go to the main site if some fail to render.In my earlier finite map...

Math typesetting with jsMath

8 months ago by Mauricio Fernandez on Eigenclass.
This entry uses jsMath to display mathematical expressions. Please go to the main site if some fail to render.I've added math typesetting...

Finite maps galore: imperative code strikes back

9 months ago by Mauricio Fernandez on Eigenclass.
Matías Giovannini has been implementing purely functional data structures based on Bentley and Sedgewick's ternary search trees (TST) and pitching them...

Making mutable state faster: optimizing the caml_modify writer barrier

9 months ago by Mauricio Fernandez on Eigenclass.
As I explained yesterday, the caml_modify write barrier needed by OCaml's incremental + generational GC is the main reason why immutable data can often...

When immutable data is faster: write barrier costs

9 months ago by Mauricio Fernandez on Eigenclass.
This example shows that in some cases immutable data can be faster than mutable state. One reason is that language implementations with incremental, generational...

Looking for a message broker; semantics of the RabbitMQ STOMP adapter

10 months ago by Mauricio Fernandez on Eigenclass.
Albeit under-specified and limited in functionality, STOMP is made attractive by its utter simplicity if you're OK with basic message queues. Yet, there...

Markdown redux, and walking the AST to generate (statically) valid XHTML

11 months ago by Mauricio Fernandez on Eigenclass.
My earlier post (about Markdown processors and how simple top-down parsers can be up to two or three orders of magnitude faster than regexp-based gsub kludges...

On the sad state of markdown processors, and getting thousandfold speed-ups.

11 months ago by Mauricio Fernandez on Eigenclass.
When I started to write the code for the latest incarnation of eigenclass, I was planning to use an existent Markdown processor to generate the HTML for...

MiniLight renderer cleanup, simpler (safer) code is faster

11 months ago by Mauricio Fernandez on Eigenclass.
MiniLight is a minimal global illumination renderer that has been translated to 7 languages and takes on average 570 lines of code. Neat! It's of course...