Microsoft Files For 3 Parallel Processing Patents

theodp writes "Microsoft may have been a Johnny-come-lately when it comes to parallel programming, but that's not stopping the software giant from trying to patent it. This week, the USPTO revealed that Microsoft has three additional parallel-processing patents pending — 1. Partitioning and Repartitioning for Data Parallel Operations, 2. Data Parallel Searching, and 3. Data Parallel Production and Consumption. Informing the USPTO that 'Software programs have been written to run sequentially since the beginning days of software development,' Microsoft adds there's been a '[recent] shift away from sequential execution toward parallel execution.' Before they grant the patents, let's hope the USPTO gets a second opinion on the novelty of Microsoft's parallel-processing patent claims."

Old stuff

[Linker3000]

Shame I don't have any of the code developed in the mid-late 80s where I worked for the Transputer - it's probably riddled with prior art.

Re:pffff

[NewbieProgrammerMan]

You'd think that somebody that's light-years ahead when it comes to parallel processing would rule the roost in the Top 500 [top500.org] supercomputer list. I'm sure there's a good explanation, though....just waiting to hear it. :)

so basically MS invented MapReduce and PVM algos

[Lazy Jones]

These patents look very much like they would cover MapReduce, the parallel search patent is obviously just a specification of a straightforward parallel search implementation e.g. using PVM (many trivial implementations exist since the 90s).

The USPTO will once again be the laughing stock of the whole IT field if they grant these patents.

Parallel database

[AlecC]

Many years ago there was a data storage sytem developed by, I think, ICL, which had a correlator built into the disk data path for each disk arm so that it could search every head of a multi-platter disk at the same. But indexing turned out to be a better method than brute force, so it died. But it was, nonetheless, a highly parallel database search.

Re:Pain of Patents is in the reading

[Anne Thwacks]

The UK's CARDS (Content Addressable Relational Data Store) database engine, on which I worked in the 1980s, did this kind of stuff (Using RAID, though we didnt call it that at the time). Data was retrieved from multiple HDs using an array of Transputers, managed by a workstation that was similar to Sun workstations of the day (double-extended triple Eurocard with Motorola 68020 processor and Unix). Data relationships described graphically (like the stuff in Access, but more powerful).

So its not, like, a novel invention or anything. Please can I have a patent on "a round device for rolling heavy loads along paved areas". Thanks.

Re:Pain of Patents is in the reading

[Anonymous Coward]

MS SQL server has, since at least V7 (sql server 7) a parallel query facility, called Intra-Query Parallelisation. It is supposed to exploit extra cores for a single query.
It is buggy. In Sql server 7 it regularly deadlocked against itself, in sql server 2000 it actually could run slower than using a single core both wall-clock time and total cpu time, and sometimes much worse things happened under heavy load.
Disable it. You can do this from enterprise manager.

scientific computation

[pigwiggle]

is where the massively parallel applications are at. I regularly write and run parallel code that will efficiently run over thousands of processors - my largest run to date was over 1024, 8 processor nodes, so 8192 processes parallelized. It is all Linux - no exception. I've yet to hear of a respectable production cluster running Windows. In fact, I have yet to run into anybody who isn't running Linux on their desktop, in my line of work. I regularly write and run parallel code - the analysis code for crunching the enormous data sets produced on the clusters - for my quad core desktop machine running Linux. You've no clue.