Forgot your password?
typodupeerror
Databases

+ - Cheap GPU Accelerated Database System Competes With Top TPC-H Scores-> 1

Submitted by Anonymous Coward
An anonymous reader writes "Alenka is an open source SQL-lite database system that leverages CUDA to offload all of the query processing to NVIDIA GPUs. Newly published results show that an $800 desktop system (with an NVIDIA GTX580, and 1 120GB SSD) outperforms the top10 ranked $55,000 HP server with 2 Quad Core (3Ghz) Xeons, 144GB of RAM, and 12 60GB SSDs running Microsoft SQL Server 2008 in one query, and offers comparable performance in another.

To achieve these results, the author (Anton) recently updated the code with a new version that includes a faster implementation of database JOIN. The code leverages the Thrust library for fast SORT, SELECT, and SET parallel algorithms. It also uses the CUDPP library to implement a parallel hash JOIN.

While the codebase is not a complete implementation of SQL, it can execute several queries from TPC-H (an industry standard data-warehousing benchmark). For Query 1 (SELECT, GROUP-BY) Alenka processes a 100GB dataset in 9.5 seconds, compared to 42.3 seconds on the HP system. For Query 3 (JOIN, GROUP-BY, SORT), Alenka takes 5.3 seconds, compared to 4.3 seconds for the HP system.

It will be interesting to see if Alenka can offer similar results for the entire TPC-H benchmark suite, or if other database implementations can be accelerated by GPUs.

The source code for the Alenka system is available on github."

Link to Original Source

Comment: Why? (Score 3, Informative) 71

by Gregory Diamos (#30541194) Attached to: An Open Source Compiler From CUDA To X86-Multicore

So there seem to be several questions as to why people would want to use CUDA when an open standard exists for the same thing (OpenCL).

Well, honestly, the reason why I wrote this was because when I started, OpenCL did not exist.

I have heard the following reasons why some people prefer CUDA over OpenCL:

  • The toolchains for OpenCL are still immature. They are getting better, but are not quite as bug-free and high performance as CUDA at this point.
  • CUDA has more desirable features. For example, CUDA supports many C++ features such as templates and classes in device code that are not part of the OpenCL specification.

Additionally I would like to see a programming model like CUDA or OpenCL replace the most widespread models in industry (threads, openmp, mpi, etc...). CUDA and OpenCL are each examples of Bulk Synchronous Parallel models, which explicitly are designed with the idea that communication latency and core count will increase over time. Although I think that it is a long shot, I would like to see more applications written in these languages so there is a migration path for developers who do not want to write specialized applications for GPUs, but can instead write an application for a CPU that can take advantage of future CPUs with multiple cores, or GPUs with a large degree of fine-grained parallelism.

Most of the codebase for Ocelot could be re-used for OpenCL. The intermediate representation for each language is very similar, with the main differences being in the runtime.

Please try to tear down these arguments, it really does help.

Programming

An Open Source Compiler From CUDA To X86-Multicore 71

Posted by timothy
from the abstraction-gains-a-layer dept.
Gregory Diamos writes "An open source project, Ocelot, has recently released a just-in-time compiler for CUDA, allowing the same programs to be run on NVIDIA GPUs or x86 CPUs and providing an alternative to OpenCL. A description of the compiler was recently posted on the NVIDIA forums. The compiler works by translating GPU instructions to LLVM and then generating native code for any LLVM target. It has been validated against over 100 CUDA applications. All of the code is available under the New BSD license."
Security

+ - Is Code Auditing Of Open Source Apps Necessary?

Submitted by Anonymous Coward
An anonymous reader writes "Following Sun Microsystems' decision to release a raft of open source applications to support its secure cloud computing strategy, companies may be wondering if they should conduct security tests of their customized open source software before deployment. whilst the use of encryption and VPNs to extend a secure bridge between a company IT resource and a private cloud facility is very positive — especially now that Amazon is best testing its pay-as-you-go private cloud facility — it's important that the underlying application code is also secure. What do you think?"

+ - An Open Source Compiler From CUDA to x86-Multicore->

Submitted by Gregory Diamos
Gregory Diamos (1706444) writes "An open source project, Ocelot, has recently released a just-in-time compiler for CUDA, allowing the same programs to be run on NVIDIA GPUs or x86 CPUs and providing an alternative to OpenCL. A description of the compiler was recently posted on the NVIDIA forums. The compiler works by translating GPU instructions to LLVM and then generating native code for any LLVM target, it has been validated against over 100 CUDA applications. All of the code is available under the New BSD license."
Link to Original Source
User Journal

Journal: RFID Killer

Journal by smitty777

Nothing big today - just an RFID Terminator Gun. It basically fries any RFID chip in range. Not sure what good it is, unless you want to play a trick on your friends and family by frying their passports. Big fun.

Pause for storage relocation.

Working...