OpenStack Nova

One of my personal highlights of the recent Barcelona Summit was a session by Mirantis engineers Elena and Oleg titled “Is OpenStack Neutron Production Ready for Large Scale Deployments?”. In the session they outline a comprehensive control and data ... [More] plane testing effort, run on two labs, one with 200 nodes and run of the mill hardware, and […] [Less]

Learn more about this new event and how a shorter cycle impacts the upcoming software release. The post How the new Project Team Gathering event affects OpenStack Ocata appeared first on OpenStack Superuser.

The post Scaling OpenStack With a Shared-Nothing Architecture appeared first on Mirantis | The Pure Play OpenStack Company. In shared-nothing architecture, a distributed system can scale without limits by removing single points of contention. Learn ... [More] about pros and cons of deploying OpenStack with a shared-nothing paradigm. The post Scaling OpenStack With a Shared-Nothing Architecture appeared first on Mirantis | The Pure Play OpenStack Company. [Less]

The post Bailian: From Brick & Mortar to Brick & Click using OpenStack, DevOps appeared first on Mirantis | The Pure Play OpenStack Company. Chinese retail giant Bailian Group transitions from brick and mortar to omni-channel commerce by ... [More] using OpenStack and DevOps for increased innovation, faster time-to-market, and efficient operations. The post Bailian: From Brick & Mortar to Brick & Click using OpenStack, DevOps appeared first on Mirantis | The Pure Play OpenStack Company. [Less]

If you want to submit a patch to fix one of these hacking issues, there's a right way, says Trove project team lead (PTL) Amrith Kumar. The post Enabling hacking extensions: The right way appeared first on OpenStack Superuser.

IBM Research & OSIC Speed-up Big Data QueriesdfinebergDecember 2, 2016As a world-renowned IT research organization, IBM produces substantial breakthroughs.  Despite their size, IBM Research maintains an environment that nurtures some of the most ... [More] innovative thinkers in the world. Yosef Moatti, from the IBM Research Haifa lab in Israel, is one of those innovative thinkers.  As the lead for the IBM contribution to IOSTACK, a Horizon 2020 European project, he worked with the OpenStack Innovation Center (OSIC) to demonstrate a dramatic improvement in both performance and throughput of SQL queries on Big Data sets. Yosef Moatti, lead for IBM contribution for the IOSTACK project Yosef and his team spent three weeks using a 65-node OSIC cluster to test performance improvements of Spark-based SQL queries retrieving data from an OpenStack swift cluster. Their goal was to find a solution to the data ingestion problems that were slowing things down. “We did experiments with up to 3TB.  We plan to experiment with even larger sets of data, but we are already seeing very good results,” Yosef said.  “Our ‘push-down’ technology takes advantage of idle CPUs in the object store to reduce and even suppress network bottlenecks between the object store cluster and the compute cluster.” He explained that the concept, called active storage, uses the storage cluster to perform part of the computation.  Unlike Hadoop, Spark was also designed for interactive computation, their tests demonstrated that Spark SQL and object stores empowered with active storage can deliver on that promise even for Big Data. “Active storage was added to Swift by the storlet project, initiated by Eran Rom, a former IBM Research colleague,” he said.  “The idea of using automated SQL filter push-down has been implemented in databases for a while, where user-defined functions are pushed down to the database to filter data at its source.  We took that idea and adopted it for object store, by enhancing the storlet with the features needed to process Big Data requests.”   According to Yosef, the benefits of storlets go way beyond this enhancement.  Aside from saving bandwidth by eliminating unnecessary data transfers, storlets can be used to introduce new services. Since storlets are dynamically loaded code, the function is only limited by the developer’s imagination. For a physician doing rounds, the storlet can send only the portion of a patient’s x-ray needed to her wireless device for immediate viewing. The push-down technology, which takes advantage of mostly-idle CPUs at the object store, most importantly, reduces the used network bandwidth but also alleviates the CPU and memory pressure on the Spark side. Specifically, the team measured an increased CPU usage of 20% for the storlet software at the object store side. This is a good reason to use powerful servers for object storage. With OSIC’s help, the IBM Research team demonstrated a 32X speed-up of SQL queries*.  “Without our push-down technology, it took more than 60 minutes to process an SQL query against 3TB of accumulated readings from IoT electrical meters.  With the push-down technology, it now takes 2 minutes. The experiments run at OSIC enabled us to submit the results as a research paper to a key Systems conference,” Yosef said.  “This also validates the features added to the storlet project, which will be an official OpenStack project as of the Ocata release.”  For more information, see the related talk at OpenStack '16 Barcelona conference, the Github OpenStack Storlets page, and the IOSTACK Project page. _________________________________________________________ * Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at intel.com. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. [Less]

Azure Service Fabric is a distributed systems platform that makes it easy to package, deploy, and manage scalable and reliable microservices. It can be deployed on Azure, but also on-premise, thus making OpenStack a great choice for the latter. We ... [More] are going to use Juju to deploy our Azure Service Fabric cluster directly on OpenStack… The post Azure Service Fabric on OpenStack appeared first on Cloudbase Solutions. [Less]

Our excellent Training & Certification team has posted some videos in our RedHatCloud youtube channel that quickly go over the installation procedure of Red Hat OpenStack Platform 8, and how to boot a CloudForms instance to perform basic ... [More] management functions. Kudos to our awesome video team (Jim Meegan and Ben Oliver) and to our curriculum […] [Less]

The QEMU Advent Calendar website 2016 features a QEMU disk image each day from 01-DEC to 24 DEC. Each day a new package becomes available for download (of format tar.xz) which contains a file describing the image (readme.txt or similar), … Continue reading →

VirtualBox is a freely available hardware virtualization software from Oracle that is widely used for creating virtual computers. With cloud computing getting more attention by both companies and individual developers, VirtualBox has become a de ... [More] facto standard tool for cloud computing development. As a developer, if you only need to occasionally create a virtual machine, the graphical user interface provided by the software is very easy to use and gives you a great deal of options for you to change the characteristics of a virtual machine (VM). The post VirtualBox Networks and Virtual Machine Auto-provisioning appeared first on IBM OpenTech. [Less]