Year: 2010

In January virtualization.info reported about a stealth startup called Pancetera.

At that time there was almost no information about the company’s technology but a brief list of provided benefits. A few additional hints found online suggested that the product, possibly called TeraCapture was a storage management solution for VMware virtual infrastructures.

The real product, now that Pancetera unlocked its website, is slightly more complex as it combines the storage management with the storage optimization.

The storage management component is called SmartView.
SmartView aggregates all the storage resources available inside the virtual infrastructure as a hierarchy, under the P: virtual drive.
It supports fibre channel and iSCSI SANs, NFS arrays and even DAS resources like local hard drives. 
Any operating system or management solution that can read and map a CIFS shared folder or a NFS volume can remotely access the P: network drive, performing files management with simple drag&drop operations.

Read more

Yesterday VMware announced its financial results for the Q2 2010. During the earnings call, the company’s executives briefly commented on the market adoption of VDI and the forecast for the near future.

VMware continues to be surprisingly skeptic about the imminent adoption of VDI as a mainstream technology. And while it’s evident that most companies are not yet sold to the idea of client virtualization, this still is a radical change in the message that the delivered in the last couple of years.

In the Q1 2010 earnings call To Nielsen, COO, said:

Exactly when this market is going to tip though we don’t know. We were saying – engaged and focused on it, but I couldn’t tell you if its going to be at the end of the year or next year or exactly when that’s going to be but our eyes are certainly on the ball and we are going to make sure that when it does tip, we are there to take advantage of it.

In this call, three months later, he reiterated:

We continue to hold high expectations for the desktop virtualization market, yet it remains difficult to predict at what pace customer interest and evaluations will turn into accelerated buying. We are seeing Windows 7 upgrades, and the proliferation of new end-user devices such as the iPad and Smart phones are fueling public discussion and customer interest. However, no single technical or economic tipping point is emerging as the accelerant to VDI adoption.

During the Q&A part of the call, Nielsen added more details:

Read more

Yesterday VMware announced its financial results for Q2 2010.

Interestingly, the company started the call reporting a record growth in the SMB segment. VMware expects to accelerate this trend even more with the inclusion of vMotion in the vSphere 4.1 Essential Plus SKU. At the moment the vSphere Essentials business is growing 100% sequentially.
VMware added 20,000 new customers in the last two quarters, reaching 190,000 customers total, but it doesn’t say how many are in the SMB segment.

The company closes the Q2 quarter with $2.8B cash and $1.5B of deferred revenue, a year-over-year growth of 33%. 
Total revenue for the Q2 was $674 million, up 48% from Q2 2009, while total license revenue was $324 million, an increase of 42% from the Q2 2009. 
Software maintenance and support revenue was $290 million, an increase of 54% from last year, but the revenue decreased sequentially. Customers continue to buy on average more than 24 months of support and maintenance with each new license purchase.
Professional Services revenue was $60 million, an increase of 54% from last year, and up 11% sequentially, but VMware doesn’t expect a strong growth in the PSO revenue because of the investments in the partners ecosystem (VKernel seems to disagree here).

Read more

Demand Technology Software is a US firm that develops a number of products for performance data collection and monitoring on Windows operating systems.

The company’s flagship product is called Performance Sentry. Its multi-tier architecture is designed to scale up to thousands of monitored servers, storing their performance data on its SQL Server database. To collect data, it uses an agent that runs as a Windows service.

The company just released a version of this product that supports VMware vSphere (both ESX and ESXi hosts, with or without vCenter Server). The new version, called Performance Sentry VM, does more than just run inside a virtual machine.

Leveraging the VMware APIs in fact, Demand Technology Software managed to translate the VMware’s performance metrics into objects and counters that Windows Performance Monitor (Perfmon) can use.
The translated data, coming from the hypervisor and the guest operating systems, is presented by the Performance Sentry virtual machine that basically runs as a proxy provider. At this point any management solution can use it, including Perfmon or System Center Operation Manager (SCOM) thanks to a Management Pack that the company develops.

Read more

In the last few days virtualization.info recommended a couple of new/refreshed, very interesting papers from VMware: The Design and Evaluation of a Practical System for Fault-Tolerant Virtual Machines and Understanding Memory Resource Management in VMware ESX 4.1.

There’s at least a third one that is worth a read: Enhanced VMware ESX 4.1 CPU Scheduler.

The 22-pages paper starts with a brief introduction on the role of the CPU scheduler and then goes on with an in-depth analysis of the design changes applied to ESX 4.1:

ESX 4.1 introduces wide-VM NUMA support, which improves memory locality for memory-intensive workloads. Based on testing with micro benchmarks, the performance benefit can be up to 11–17 percent.

In ESX 4, many improvements have been introduced in the CPU scheduler. This includes further relaxed co-scheduling, lower lock contention, and multicore-aware load balancing. Co-scheduling overhead has been further reduced by the accurate measurement of the co-scheduling skew and by allowing more scheduling choices. Lower lock contention is achieved by replacing the scheduler cell lock with finer-grained locks. By eliminating the scheduler cell, a virtual machine can get higher aggregated cache capacity and memory bandwidth. Lastly, multicore-aware load balancing achieves high CPU utilization while minimizing the cost of migrations.

It also includes an entire benchmark section with the performance evaluation of the updated design, comparing ESX 3.5 with ESX 4.

Spiceworks is a private US-based company founded in 2006 that offers a completely free system management platform for hardware/software/licenses inventory, change management, helpdesk, OS remote control and network mapping.

The company introduced support for VMware vSphere in version 4.6, released in March.

Now it’s the turn of Hyper-V: the upcoming release 5.0 will include support for the Microsoft hypervisor according to an official communication published earlier today by the company:

• See all the virtual machine data on MS (& VMware) virtual servers as well as details like allocated RAM and hard disk space.
• Start and stop your VMs when the need arises – allowing for more sophisticated virtualization management.
• See and report changes over time for a VMs state and configuration
• See all your VMs across all your hosts

The major update is planned for September. 
Well before that date Spiceworks 5.0 will be available as a public beta.

Just a couple of weeks ago Microsoft opened the public beta program for the Windows Server 2008 R2 and Windows 7 Service Pack 1.

The update is especially interesting as it introduces two key technologies for virtualization: Dynamic Memory for Hyper-V and RemoteFX for Remote Desktop Services (RDS).
Earlier this year, widespread rumors suggested that the SP1 would arrive within the end of this year, but Microsoft recently published in its FAQ page, a different roadmap which places the update somewhere in H1 2011.
Mary Jo Foley at ZDNet is reporting the rumored date of April 2011.

Microsoft is trying to change the market perception about its capability to execute, fighting its fame of over-promising and under-delivering vendor, while surprising competitors at the same time. So it won’t be too surprising to see the SP1 indeed arriving before the end of this year.

In February, CA hired Andi Mann, the former Vice President of Research at Enterprise Management Associates (EMA) in February.
In his new role as Vice President of Product Marketing for Virtualization Management, Mann helped to reorganize the go-to-virtualization-market strategy.
virtualization.info also received a tip that CA hired Mike O’Malley, the former Director of Analyst Relations and Market Intelligence at VMware.
O’Malley is the Vice President of Product Marketing since January.

The two guys probably worked together at a plan to refresh the company’s image and credibility as a virtualization player.
The process culminated with the relaunch of the CA’s virtualization portfolio, which now includes five products, officially released yesterday:

Read more

Like most VMware partners, PHD Virtual rushed to release an update for its flagship product, Backup for VMware (formerly esXpress), that supports the recently released vSphere 4.1 platform.

Announced yesterday, version 4.0-4 does exactly that, as well as enhancing the product in several different areas, and it’s shipping today, available for 4.0 customers at no cost.

In May VMware published a very interesting paper describing the design of the Fault Tolerant (FT) feature, announced for the first time in late 2007 and shipped with vSphere 4.0 in June 2009.

The document, titled The Design and Evaluation of a Practical System for Fault-Tolerant Virtual Machines, also describes the alternative designs that VMware explored before selecting the actual one::

We have implemented a commercial enterprise-grade system for providing fault-tolerant virtual machines, based on the approach of replicating the execution of a primary virtual machine (VM) via a backup virtual machine on another server. We have designed a complete system in VMware vSphere 4.0 that is easy to use, runs on commodity servers, and typically reduces performance of real applications by less than 10%. Our method for replicating VM execution is similar to that described in Bressoud, but we have made a number of significant design changes that greatly improve performance. In addition, an easy-to- use, commercial system that automatically restores redundancy after failure requires many additional components beyond replicated VM execution. We have designed and implemented these extra components and addressed many practical issues encountered in supporting VMs running enterprise applications. In this paper, we describe our basic design, discuss alternate design choices and a number of the implementation details, and provide an evaluation of our performance for both micro-benchmarks and real applications.

Thanks to Yellow Bricks for the news.