Virtualization.com

News and insights from the vibrant world of virtualization and cloud computing

virtualization

KVM Forum 2008 Schedule Has Been Announced

by 1 Comment

The schedule for the upcoming KVM 2008 forum in Nappa, CA is up.

Qumranet, as the main KVM sponsor, is inviting all KVM developers to their second KVM summit on June 10 to 13th at the Marriot Napa Valley, California.

Many of the world’s top kernel developers will gather to discuss the state of the union on KVM and virtualization technology in general. More specifically, the group will plan the technology roadmap and future of KVM.

Avi Kivity will be keynoting, and off course Qumranet will also talk about KVM in Solid ICE. Apart from that, there will be a variety of presentations from Red Hat, IBM, Transitive and Intel representatives.

Gerd Hoffmann of Red Hat (SUSE in a previous role) will be talking about mixing Xen and KVM with xenner, which is a utility able to run Xen paravirtualized kernels as guests on Linux hosts, without the Xen hypervisor and using kvm instead.

Different IBM people will be discussing the state of KVM on Big Iron and PowerPC. And there will also be some talk about Open-ovf , an open source software project around the Open Virtual Appliance Format.

Stay tuned for more!

(Full disclosure: Virtualization.com is a media partner of the KVM Forum 2008)

Xen 3.2.1 Released

by Leave a Comment

Keir Fraser has announced the availability of Xen 3.2.1 and 3.1.4, both bugfix releases in the Xen 3.1 and 3.2 series.

The 3.2.1 release contains architectural improvements and new user-visible features including:

  • Xen Security Modules (XSM)
  • ACPI S3 suspend-to-RAM support for the host system
  • Preliminary PCI pass-through support (using appropriate Intel or AMD I/O-virtualization hardware)
  • Preliminary support for a wider range of bootloaders in fully virtualized (HVM)guests, using full emulation of x86 real mode
  • Faster emulation of standard (non-super) VGA modes for HVM guests
  • Configurable timer modes for HVM guests, depending on how the guest OS manages time-keeping
  • Many other changes and enhancements across all supported machine architectures

As usual downloads are available at Xen.org

Novell CTO Jeff Jaffe Outlines Strategic Roadmap (Including Virtualization)

by Leave a Comment

Novell CTO Jeff Jaffe published a blog post earlier today outlining the company’s roadmap in seven areas: policy, identity, virtualization, Linux, orchestration, compliance and collaboration. Of course, the virtualization one interested us the most:

  • In 2006, Novell shipped Xen as an integrated part of our SUSE Linux Enterprise platform. In 2007, we collaborated with Microsoft to deliver the first cross-platform solution for running Windows on Linux.
  • At BrainShare, for the first time anywhere in the world, we demonstrated live migration of Windows Server 2008. We had it running as a Xen virtual machine on SUSE Linux Enterprise Server.
  • Our vision for virtualization is that the p-Distro becomes the core operating system for the physical machine and hosts the v-Distros. To get here, we have work to do: performance tuning, ISV certification, systems management, security improvements and device drivers.

Read the rest here.

Parallels Named Red Herring 100 Award Finalist

by Leave a Comment

Server virtualization firm Parallels has been named a finalist in the Red Herring 100 Awards‘ selection of the most innovative private technology companies based in North America.

Red Herring

In preparation for the formal announcement of its Top 100 Award, the Red Herring editorial board surveyed the entrepreneurial scene throughout the North American region and identified the top 200 out of more than 800 closely evaluated companies that are “leading the next wave of innovation”.

Evaluations were made on both quantitative and qualitative criteria such as financial performance, innovation, management, strategy, and ecosystem integration.

Joel Dreyfuss, Editor-in-Chief of Red Herring noted:

”We can see the exciting evolution of the technology sector reflected in the quality and variety of exceptional companies that we had to choose from in putting our list together. It was tough to choose just 200 finalists from such a large list of excellent contenders, and we are very happy with the quality of the companies we selected as finalists.”

According to Red Herring, Parallels is on the list due to its innovative and entrepreneurial achievements. The 100 winning companies will be announced at the Red Herring North America event, May 12-14, in San Jose.

[Source: The Hosting News]

Nortel Spinoff Blade Network Technologies Aims To Virtualize Datacenter Racks

by Leave a Comment

BLADE Network Technologies, the data-center server-switch company spun off from Nortel two years ago, announces the availability of its new RackSwitch network virtualization switches—the only switches designed specifically for IBM’s iDataPlex, an inexpensive, custom-configured rack system featuring design innovations in cooling and efficiency to address Web 2.0-style computing (see iDataPlex coverage on GigaOM and Ars Technica).

BLADE Network Technologies

From the press release:

RackSwitch provides the high-bandwidth communications for IBM iDataPlex in today’s massive scale-out data centers that run I/O-intensive Web 2.0 applications and seek to make the most efficient use of the cloud computing model. Leveraging IBM and BLADE’s common blade server heritage, RackSwitch enables iDataPlex to increase the density networking within a single rack, use significantly less power for networking while maximizing bandwidth available to a single system, employ server-friendly cooling of the networking subsystem and provide 100-percent interoperability with existing network infrastructures. BLADE’s RackSwitch delivers standardized networking with the lowest latency and highest throughput, line-rate/non-blocking switching performance at 1 Gigabit and 10 Gigabit, with stacking and unified management.

Blade claims leadership in this particular market, having installed more than four million ports connecting more than 800,000 HP, IBM and NEC server blades, with products deployed across 26 market segments.

BLADE’s RackSwitch G8100, a 1U top-of-rack switch equipped with 24 lossless, low-latency 10 Gigabit Ethernet (10 GE) ports, is designed to equip iDataPlex in high-performance clusters that require 10 Gigabit Ethernet bandwidth with latency of 300 nanoseconds or less. BLADE’s RackSwitch G8000, a 1U top-of-rack switch equipped with 48 Gigabit Ethernet ports and four 10 Gigabit Ethernet ports for uplinks and/or stacking, is designed for the use of iDataPlex for emerging high-volume cloud computing environments.

BLADE’s new RackSwitch products are available for iDataPlex in the US and Canada in June and globally by the end of the year at a starting price of US$5,500. The company will demonstrate the new switches at next week’s Interop conference in Las Vegas.

[Source: New York Times / NetworkWorld]

Sysload Software Releases SP-ANALYST, Performance Management Solution for VMware ESX

by Leave a Comment

Sysload Software, a provider of performance and capacity management solutions for data centres, announced today that it has released SP-ANALYST for VMware ESX infrastructures.

Sysload

SP-ANALYST for VMware ESX provides increased frequency and higher granularity of data collection, the use of a single console to monitor both physical and virtual machines, and an extra feature that controls the timekeeping discrepancy on machines running the ESX Server platform. By providing a centralized, real-time view of the data centre, Sysload’s solutions aim to make performance management easier, even on the most complex virtual infrastructures.

From the press release:

Sysload’s full range of solutions supports multi-platform environments in terms of equipment, operating systems, and proprietary virtualization platforms (e.g., VMware, SUN Solaris, HP-ux, IBM Aix, IBM i5/OS, Linux, Windows, Novell, Citrix XenServer).

Sysload’s data collection technology is based on dedicated data collection agents deployed directly on the monitored system that allow objective monitoring of an ensemble of physical and/or virtual servers. Sysload agents can collect up to 300 metrics directly from the heart of the monitored systems with very high frequency (intervals as low as 1 second) while consuming minimal resources (less than one percent of CPU with no continuous network traffic thanks to localized history files). This fine data granularity ensures precise monitoring of system utilization and allows the detection of virtualization-related micro saturation incidents that cannot be detected by most other monitoring tools on the market.

[Source: Sysload pressroom]