Feed on
Posts
Comments

Intel has unveiled a new multicore processor based on technology from Larrabee, the advanced graphics chip that was placed on hold late last year.

The new chip is codenamed Knights Corner and is based on Intel’s Many Integrated Core (MIC) architecture. The processor is set scale to more than 50 processing cores.

Intel described the MIC architecture as being based on prior Intel projects which include Larrabee and other Labs research such as the single-chip cloud computer. The MIC architecture is designed for highly parallel architecture and is separate from the Intel Xeon chips that are used for their main consumer and business product range.

GigaSpaces have added Memcached protocol support in version 7.1 of GigaSpaces XAP.  memcached is a popular caching protocol and server. The integration provided allows to use GigaSpaces as a memcached server (talking the memcached protocol) enhancing the memcached to provide high availability, write behind, and other features that are provided by GigaSpaces XAP.

Further details can be found here.

The VisitMix website has some very nice tips and tricks for accessing and writing to Windows Azure Blog storage. The post goes into detail on:

- Creating An In Memory .xls File And Uploading It To Blob Storage

- Creating and Unzipping a .Zip File From A MemoryStream In Azure

- Using WebClient To Access Blobs vs. CloudBlobClient– Benchmarking Time

MIT via their OpenCourseware portal are giving access to their lecture notes and video for their multicore programming primer course. This is an excellent resource for those interested in multicore programming.

Like many people I have files stored on a number of different place on the ‘cloud’. Some are on email (several different providers), and others are on free storage sites, some are on FTP, documents are on Google Docs etc. Accessing these files as and when I need them is not hard if I’m sat in front of my PC, but the fact they are distributed in different services does present some challenges.

I’ve been using SMEStorage for a while now to tie these services together into a ‘virtual cloud file system’ so that I can see and manage them all from one place. It has worked pretty well for me and with the release of new clients, that include a linux cloud drive and Mac Integration it has made it more applicable to the devices and OS’s I use everyday.

Given I spend a lot of time on the move, what was missing was mobile access, in particular iPhone access. Happily this has now been resolved with iSMEStorage now in the App Store.

Almost everything you can do with the web version you can do on the phone version, which includes:

Features include:

- File Viewer

- DropBox like Sync but for your cloud,

- Memo / Voice Memo creation and sync, Ability to load Local Files,

- Secure Files that need password before able to view,

- Cloud Clipboard: which enables you to share file links over email for multiple files from multiple storage clouds

- Get a TinyURL for any file posted to the iPhone clipboard for use in other applications,

- Create / Manage Collaboration Groups for sharing files

- If you want to Sync your files in real time with your underlying Storage Cloud, you can initiate this from Cloud Sync
and it is carried out on the Server.

- Easily rename and upload Photos to your chosen storage cloud

- Cloud Providers: Manage your cloud providers in real time and change which one you want files to be uploaded to
from the iPhone.


A video overview of the App’ can seen below:





Clouds supported include:

- Amazon S3
- RackSpace Cloud Files
- Box.Net
- Google Docs
- Microsoft SkyDrive
- Microsoft Live Mesh
- GMail-as-a-Cloud
- Email Clouds
- FTP Clouds
- Apple MobileMe
- WebDav enabled clouds

Sun (or is that Oracle…) has released a new version of their Grid Engine which brings it into the cloud.

There are two main additions in this release. The First is is integration with Apache Hadoop in which Hadoop jobs can now be submitted to Grid Engine, as if they were any other computation job. The Grid Engine also understand Hadoop’s global file systems which means that the Grid Engine is able to send work to the correct part of the cluster (data affinity).

The second is dynamic resource reallocation which also includes the ability to use on-demand resources from Amazon EC2. Grid Engine also is now able to manage resources across logical clusters which can be either in Cloud or off Cloud. This means that Grid engine can now be configured to “cloud burst” dependent on load which is a great feature. Integration is specifically set up with EC2 and enables scale down as well as scale up.

This release of Grid Engine also implements a usage accounting and billing feature called ARCo, making it truly SaaS ready as it is able to cost and bill jobs.

Impressive and useful stuff, and if you are interested in finding out more you can do so here.

Further to the post that we outlined from Ricky Ho on NOSQL, he now has an additional post on Query Processing for NOSQL. This is well worth a read as many of the NOSQL products have fairly primitive query capabilities and Ricky outlines how this can be approached. I’d recommend reading the comments at the foot of the blog which  provide some good reading on discussion around the approach.

It seems that Apple has pulled the project to implement the ZFS file system over the mac, apparently due to licensing issues (no surprise there, now that Sun is in the hands of Oracle).


Mac’s OX extended format is synonymous with HFS+. HFS+ used B-Trees to store volume metadata and has being around since Mac OS 8.1 so the news that Apple were considering ZFS created quite a buzz at the time.


The Mac OS had limited ZFS support with Leopard and there was a hope that there were would be full support in Snow Leopard, but it was not to be. 


ZFS fundamentally changes the storage equation by integrating devices,storage, and file system structures into a single structure. By integrating the file system with volume management, the risk of misconfigurations at one layer affecting another layer is virtually eliminated.


Meanwhile Apple has advertised openings for File Systems engineers so it appears they intend to move forward in another direction.

Ricky Ho has done a great job of providing a thorough overview of the characteristics and patterns of what are being terms NOSQL products. This includes looking at products purporting to be NOSQL products, API’s ,data partitioning, data replication, client consistency, vector clock, and more. A very good read.


Interesting Post from Perils of Parallel which outlines Tim Sweeney, found of Epic Games,  keynote at High Performance Graphics 2009 in which, in one of the slides (79), he compares complexity and cost of development. Bottom line according to Tim is:

 

Lessons learned: Today’s hardware is too hard!

  • If it costs X (time, money, pain) to develop an efficient single-threaded algorithm, then…
    • Multithreaded version costs 2X
    • PlayStation 3 Cell version costs 5X
    • Current “GPGPU” version costs: 10X or more
  • Over 2X is uneconomical for most software companies!
  • This is an argument against:
    • Hardware that requires difficult programming techniques
    • Non-unified memory architectures
    • Limited “GPGPU” programming models
    Original Post here.

Older Posts »