Entries for Giovanni Marchetti

System Resource Manager

gmarchetti — Wed, 22 Oct 2008 20:18:00 GMT

This screencasts shows you how to set up and use Windows System Resource Manager. WSRM is a tool that allows you to assign CPU and memory to processes or groups of users, thus optimizing resource allocation within a machine (real or virtual).

Green S+S Dynamic Live IT ‐ Stop Climate Change Now???!!! [Green S+S Dynamic Live IT ‐ Stop Climate Change Now???!!!]

gmarchetti — Thu, 17 Jul 2008 16:53:00 GMT

“Green IT” is a catch‐all name for a series of initiatives, all aiming ultimately to save money with
the excuse of (or by means of) reducing energy consumption by datacenters, hence their
“carbon footprint”. I can think of several questions to understand how “green” we are. As for
the answers, I need time to find them.

1. What and where do you measure?
OEMs typically expose power consumption data via their management cards (where they are
available), but those data are limited to the server in question. As its workload varies, its
consumption will vary. Also, there is a “minimum” power consumption needed to keep the
server running. Although this has decreased significantly in modern machines with ACPI, power
stepping circuitry, it is still a pretty high percentage of the theoretical maximum. Switches,
racks and “environmental” systems (lights, a/c) also consume power, but few of them actually
have a way to tell how much. So, where do we measure and how do we aggregate and analyze
the measurements?

2. Have you got the appropriate hardware for the job?
For instance, blades have become very popular because of their reduced footprint. Most of
their power is allocated to the CPUs (obviously, as their is little else on the blade). For CPUintensive
applications, that is not a problem. If you also want to run data‐intensive applications,
you will likely have to store those data elsewhere, maybe on some sort of SAN. Storage systems
typically have no way of knowing what kind of application uses the data and when, hence they
keep their disks spinning all the time. From a power utilization point of view, in this case it is
more efficient to keep the data local to the system, hence a blade may not be the best solution.

3. Is your software making optimal use of such hardware?
Your workload can be optimally distributed on the hardware to minimize power consumption
and make best use of what is available. Let us suppose that you have a memory‐intensive
application on a NUMA (e.g. Opteron) machine. For best performance, you will try and allocate
1 thread of such application per CPU socket, thus optimizing the use of available memory
bandwidth. This leaves the other cores on those sockets idle. On Windows today you cannot
turn them off; active power management (processor power states, if enabled in the bios, are
supported by server 2008) will slow them down.

4. Further reading
This list of questions is definitely not exhaustive. As I look for answers, more will probably come
to mind and I’ll make a new “feature of the week” out of it. In the meanwhile, feel free to send
me your comments.
Here are a few pointers that may help inform a discussion:

- Lewis Curtis’s blog

- Little Miss Enviro-Geek

- The Green Datacenter Blog

- Microsoft’s Environment web page

- IBM Green Datacenter paper

- Green Computing Paper

- The Green Grid

in reply to Green S+S Dynamic Live IT ‐ Stop Climate Change Now???!!!

HPC Server 2008 High Availability - Part 4 [HPC Server 2008 High Availability - Part 4]

gmarchetti — Tue, 08 Jul 2008 15:23:00 GMT

This is the last of a series of screencasts demonstrating how to set up HPC Server 2008 head nodes for high availability.
In this part we discuss the installation and configuration of HPC Pack 2008 on a fail-over cluster.

in reply to HPC Server 2008 High Availability - Part 4

HPC Server 2008 High Availability - Part 3 [HPC Server 2008 High Availability - Part 3]

gmarchetti — Tue, 24 Jun 2008 07:01:00 GMT

This is the third of a series of screencasts demonstrating how to set up HPC Server 2008 head nodes for high availability.
In this part we discuss the installation of SQL Server 2005 on a 2008 fail-over cluster.

in reply to HPC Server 2008 High Availability - Part 3

HPC Server 2008 High Availability - Part 2 [HPC Server 2008 High Availability - Part 2]

gmarchetti — Tue, 17 Jun 2008 07:01:00 GMT

This is the second of a series of screencasts demonstrating how to set up HPC Server 2008 head nodes for high availability.
In this part we discuss the creation of a 2008 fail-over cluster.

in reply to HPC Server 2008 High Availability - Part 2

HPC Server 2008 High Availability - Part 1 [HPC Server 2008 High Availability - Part 1]

gmarchetti — Fri, 30 May 2008 15:19:00 GMT

This is the first of a series of screencasts demonstrating how to set up HPC Server 2008 head nodes for high availability.
In this part we discuss the creation of a shared iSCSI storage device.

in reply to HPC Server 2008 High Availability - Part 1

HPC in Media [HPC in Media]

gmarchetti — Thu, 24 Apr 2008 14:42:00 GMT

In my last post, I explained how HPC can be used to build UFOs. I have since learned about using HPC to make movies. I’d never have guessed, so let me share my surprise here.

Digital media production follows a complex workflows, from initial sketch to wireframe model, to rendered 3D images, to movies. HPC is typically used in rendering, encoding or transcoding.

Although Microsoft offers a solution to manage the workflow (Interactive Media Manager) and to encode (windows media encoder, expression encoder), we rely on partners for the rendering, most transcoding and non-windows media formats.

Rendering

Rendering is the process of transforming a 3-D model of an object (often displayed as a wireframe) into a 2-D image. Traditional non-interactive rendering (e.g. by ray-tracing) is a good example of an embarrassingly parallel problem. Several frames can be rendered at once, independently, one per computing node. The rendering software often also uses multi-threading on the node to compute several channels at once (e.g. diffuse light, reflected light, occlusions, refraction), then it composites those channels to generate the final image.

Partners that are active in this space are:

- Autodesk with Maya

- Softimage with XSI / Ray3

- Mental Images with Mental Ray

- Nvidia with Gelato Pro

All of those applications exist for Windows Server 2003 and are being ported to Server 2008. They offer varying levels of integration with our compute cluster scheduler. With built a simple demonstration with Softimage, for instance. We submitted a job with 25 frames to render on a small cluster with 1 head node and 3 compute nodes, each with 4 cores. The job was created as a parametric sweep with each task invoking the ray-tracing software installed on the node (Ray 3) on one of those frames. Ray 3 is multi-threaded and can render 8 channels, so we gave each task the full 4 cores per node.

Here’s what the rendered frame 1 looks like:

Here’s an example of one of the channels computed for the image above, i.e. the intensity of reflected light:

Encoding

Encoding a digital movie file (wmv, mpeg2 for instance) is the process of exploiting the redundancy in the sequence of frames making up that movie in such a way that fewer bits are necessary to represent them than in the uncompressed stream.

There are two kinds of redundancy:

- spatial, due to the correlation of neighboring pixels.

- temporal, due to the correlation of subsequent frames.

Thus, we could split a sequence of frames into a finite number of sections, then assign each section for computation to a separate node. We could further split each frame in the section into a number of tiles and distribute the tiles amongst cores within the node for encoding.

Microsoft has developed a VC-1 encoder (used for BluRay DVDs amongst others) and an SDK for it that enable parallel encoding. They are licensed on a commercial basis to ISVs and OEMs. An example of use is in Sonic’s Parallel Stream Encoder.

Another popular encoding format is MPEG2 (DVD, digital TV), for which several open source encoder implementations exist (Berkeley, Cornell).

The H264 standard is also widely supported and parallel encoder implementations are freely available for it (e.g. ELDER)

in reply to HPC in Media

HPC and UFOs Explained! [HPC and UFOs Explained!]

gmarchetti — Fri, 21 Mar 2008 07:01:00 GMT

I had several discussions about High Performance Computing over the past week or so with people in different job roles.
Most of them wondered: “What is HPC for and where does Microsoft fit into it?”. I thought I’d spend some time trying to answer it. Here we go:

HPC is for building UFOs

Now that Iʼve got your attention, let me try and explain: I would like to take a field of application - namely aeronautics - and take you through the design process that engineers in that field follow. Iʼll then try and explain where Microsoft technologies fit in that process. I am no expert in aeronautics - itʼs just a personal interest of mine; so, if you find any mistakes or just think I have smoked one too many, feel free to comment!

One of the hot topics in aeronautical engineering is how to overcome the design limitations of traditional airplanes. If you think about it, most commercial planes today look alike: a big pipe with arrow-shaped wings attached to it. This design is reaching its scalability limits: it is very difficult to make bigger planes that carry more people over longer distances, consuming less fuel than the current models. The recent problems and delays with both the Airbus A380, A350 and the Boeing 787 are just an example of this. One of the solutions being studied is a so called blended wing-body (BWB) aircraft.
Hereʼs an example of what Iʼm talking about:

Figure 1: An example of Blended Wing-Body aircraft.

Figure 2: A Manta ray, Natureʼs blended wing-body design

Aircraft designers typically use a variety of CAD / CAE programs to study the aircraft geometry on their workstations. A good example is IBM's CATIA.

The geometry thus generated is then discretized for finite-element analysis. This step is often called pre-processing or mesh generation. Material properties and boundary conditions (loads, constraints) are associated to the mesh elements. A variety of mathematical models are produced to study structural stress, pressure distribution, heat distribution, lift & drag, etc...

Figure 3: Example of finite-element meshes.

Software like Ansys is often used for mesh generation. Several engineering outfits write their own mesh generators as well. These packages may take advantage of parallel computing.

The models thus created are then computed by separate packages (or modules within the same commercial package) called solvers or processors. These applications definitely benefit from parallel computing and are often written for high-performance computing clusters. To put it simply, the meshes are partitioned amongst computing nodes. Each of those solves equations on the elements that pertain to it, then communicates with the neighboring nodes over low-latency links to pass boundary results. Once every node has finished, we have an overall picture of the approximated solution. Several solvers are available in the public domain. Again, part of the “secret sauce” of engineering companies is in their own solvers, their algorithms, speed and precision.

The last step of the simulation is visualization or post-processing. The results of the computation are collated and displayed in a human-intelligible form. Here's an example of it:

Figure 4: Pressure distribution over a BWB glider

The visualization step is as important as the computation itself, because it is on the visualized data that design decisions are made. The visualization process itself may benefit from parallel processing, depending largely on the quantity of data at hand.
Figure 5 summarizes the process.

Figure 5: FEM Process

What has Microsoft got to do with it?
Well, some answers are evident: We provide the operating system where the geometry designand visualization will most likely occur.

Our platform, though, can do much more than that.

First of all, the data-handling requirements of such process are huge. Not only the file sizes involved are typically in order of gigabytes, but also those files do not mean much unless the design and simulation parameters that were used are associated to them.
Here's an immediate opportunity to build a data and metadata repository centered around SQL Server. Engineers must be able to retrieve past simulation data to work on statistical models. Regulators impose that design documentation be stored for the life of the aircraft, just in case something goes wrong and decisions must be re-visited and corrected.

Secondly, the workflow that I have presented is rather simplistic and is part of a much more complex design process. Here's an example:

Figure 6: Structural, Aerodynamic and Aeroelastic Design Concept Model

Microsoft's Workflow Foundation (WF) offers the building blocks for the application logic that automates such complex workflows. It also exposes several services, like persistence and tracking, that are extremely useful to manage the complexity of such processes, making interaction with them much easier. With Visual Studio one can construct such application logic workflows and then host them as services on Windows Server. The communication amongst those services can be handled by Windows
Communication Foundation (WCF).

The compute cluster per se can also be conceived as a “computing” service exposed to the application logic via a WCF interface. In fact, thatʼs what weʼre building into HPC Server 2008.

Thirdly, HPC Server 2008 offers a computing platform for your solvers that is easy to integrate in your engineering workflows, thanks to the variety of interfaces it exposes. Its management can be mostly demanded to a traditional IT organization, thus leaving the engineers free to focus on the design. It is a relatively small and conceptually simple part of the overall solution.

Last but not least, Windows Presentation Foundation (WPF) is a powerful tool to build data visualizers relatively easily.
Sharepoint Server can be used as a presentation layer and as a WF host, masking the complexity of the underlying system. It has the advantage of enabling collaboration on design decisions as well. For instance, an engineer could use Office to produce a report on his latest simulation results, embedding links to a Sharepoint applet that
interfaces with the simulation system. Another engineer could then read the report and dig into the simulation data and parameters. If not satisfied, he could trigger another simulation from the portal and embed his results in the document as a comment. The document would then be checked back into Sharepoint.

Figure 7: Technical architecture of a computing solution

Summary
I hope that I haven't bored you too much. The morale of the story is relatively simple: Microsoft has a pretty complete platform, which you can use for high performance computing now. Projects may be long and complex compared to traditional Microsoft engagements, but their impact is huge - quite literally :-).

Figure 8: The B2 bomber.

References and Acknowledgements
Here are the sources that I have shamelessly plagiarized. A big thank you to:
C. Osterheld, W. Heinze, P. Horst, Preliminary Design of a Blended Wing Body Configuration using the Design Tool PrADO
M. Barton, S.D. Rajan, A Finite Element Primer for Engineers
Marc Holmes, Simon Cox, Delivering End-to-End High-Productivity Computing, Microsoft Architect Journal n.11

in reply to HPC and UFOs Explained!

Hyper-V & Server 2008 RTM [Hyper-V & Server 2008 RTM]

gmarchetti — Thu, 21 Feb 2008 08:01:00 GMT

With general availability of Windows Server 2008 Microsoft ships the beta 1 release of Windows Virtualization or Hyper-V (aka. Viridian).

Customers, partners, developers, IT Professionals, everyone buying a copy of Windows Server 2008 will have the opportunity to test drive the Microsoft hypervisor Hyper-V.

The version of Hyper-V included in the RTM build of Windows Server 2008 is identical to the build made available as part of the RC1 release of the server OS. The final release of Hyper-V will be available with 180 days from now. Microsoft’s hypervisor will only be available for x64 hardware platforms supporting hardware virtualization (Intel VT or AMD-V technologies). It will not be available for IA64 (Itanium) platforms. See resources for more information or here for general information about virtualization.

Hyper-V is still in beta. Support for Hyper-V is only available via online forums and newsgroups. See resources section.

The beta release of Hyper-V has limited guest operating support:

Windows Server 2008 32-bit and 64-bit (x64)

Windows Server 2003 32-bit and 64-bit (x64)

SUSE LINUX Enterprise Server 10 with SP1

Other OS installations might run as well but are not “supported” in the beta. The full list of supported operating systems in will be announced prior to RTM.

Virtualization solutions from Microsoft are:

Presentation Virtualization via Terminal Services

Application Virtualization via SoftGrid

Virtual PC

Virtual Server

Hyper-V Server Virtualization

Note:

The Virtual Server product will continue to be available. Hyper-V is only supported on 64-bit platforms, Virtual Server is the Microsoft offering for 32-bit platforms and other systems not providing the necessary hardware infrastructure for Hyper-V.

IT PROs

How does the general availability Hyper-V beta impact IT PROs? For some of the partners this is the first time they will be exposed to Hyper-V. They may be experienced with Virtual Server and/or Virtual PC. Hyper-V is a novel concept: a layer between the hardware and a variety of operating systems handling scheduling and memory allocation (amongst other things). It is the foundation of a dynamic infrastructure, making resource optimization and re-configuration far simpler than before and simplifying operations. Depending on the type of application there might be areas where its performance might be impacted. Heavily I/O bound and graphics-intensive applications are among those. Deciding where, what and how to virtualize infrastructure requires careful thought. Our Infrastructure Planning and Design series of documents will help guide that process.

Resources

Some resources you might find useful:

Forums

Microsoft SoftGrid Application Virtualization
System Center Virtual Machine Manager

Server Virtualization

Documentation

Technet Center for Virtualization

API

WMI

Hyper-V WMI API

HyperCall API

Products

Hyper-V
Virtual Server
Virtual PC
Terminal Services
SoftGrid

Programs

VHD Test Drive

Blogs

Virtualization team blog

Virtual PC Guy

John Howard

Hardware

Intel VT (aka. Vanderpool)

AMD-V (aka. Pacifica)

Miscellaneous

Virtualization Homepage

Clustering Hyper-V Screencast

in reply to Hyper-V & Server 2008 RTM

Using Powershell in HPC Server 2008 [Using Powershell in HPC Server 2008]

gmarchetti — Tue, 19 Feb 2008 08:01:00 GMT

This short video shows a few examples of how to use Powershell with HPC Server 2008 beta 1 to handle jobs. It also demonstrates exercise 7 of the HPC Server 2008 lab document previously mentioned here.

in reply to Using Powershell in HPC Server 2008

Compute Node Deployment with HPC Server 2008 [Compute Node Deployment with HPC Server 2008]

gmarchetti — Mon, 18 Feb 2008 08:01:00 GMT

This short video shows an example of how to deploy compute nodes of a HPC Server 2008 beta 1 cluster. It corresponds to exercise 2 and 3 of the HPC Server 2008 lab document previously mentioned here.

in reply to Compute Node Deployment with HPC Server 2008

Interview with Elden, Cluster PM [Interview with Elden, Cluster PM]

gmarchetti — Fri, 15 Feb 2008 07:59:00 GMT

We interview Elden Christensen, program manager for fail-over clustering, to understand what is new and improved in server 2008. He talks about clustering and geo-clustering for the rest of us. He and others have also recorded a series of webcasts on this topic, which you can find here.

in reply to Interview with Elden, Cluster PM

Job Submission and Management in HPC Server 2008 [Job Submission and Management in HPC Server 2008]

gmarchetti — Tue, 12 Feb 2008 08:01:00 GMT

This short video shows an example of how to submit jobs and monitor their status on a HPC Server 2008 beta 1 cluster. It corresponds to exercise 4 of the HPC Server 2008 lab document previously mentioned here.

in reply to Job Submission and Management in HPC Server 2008

Deploying HPC Server 2008 head node [Deploying HPC Server 2008 head node]

gmarchetti — Mon, 11 Feb 2008 08:01:00 GMT

This short video shows an example of how to set up a HPC Server 2008 beta 1 head node. It corresponds to exercise 1 of the HPC Server 2008 lab document previously mentioned here.

in reply to Deploying HPC Server 2008 head node

Windows Server 2008 Clusters [Windows Server 2008 Clusters]

gmarchetti — Fri, 25 Jan 2008 06:07:00 GMT

It is no secret that HPC Server 2008 will offer the option to make the head node of a HPC cluster highly available. This feature is not in beta 1, but it is being developed. It will exploit fail-over mechanisms provided by Server 2008 (enterprise edition or better), so I thought I'd mention some highlights in this area too.

High-availability clusters are difficult to set up and troubleshoot on several platforms. With Windows Server 2003 we made progress in simplifying them, but limitations are still significant:

You need a configuration that is fully and specifically certified as a cluster in order to obtain support when things go wrong.
There is very limited support for geo-clusters, because of limitations in intra-cluster communications, no awareness of storage location and cluster quorum models. Also, geo-clusters require yet another level of certification.
Writing cluster-aware applications is not easy. It requires knowledge of cluster-specific APIs in order to produce “resources” usable by the cluster software. Scripting generic application fail-over is supported, but limited in functionality.
Troubleshooting by reading cluster logs requires very deep knowledge to interpret the cryptic messages therein.

These limitations may hamper adoption, especially in such environments - like HPC - where Windows has not been popular.

Windows Server 2008 introduces some significant improvements that address most of those issues:

1. Configuration validation
A test tool is built into the product. It will analyze nodes and shared storage (if any) before they join a cluster. It can also be used as a troubleshooting tool, as long as the storage you want to analyze is offline. The tool will point out any issues with the hardware and the configuration that may make them unsuitable for a fail-over cluster. It will finally replace the cluster HCL. So, if the hardware passes validation, then the configuration is officially supported.

2. Simplified resource setup
A wizard-driven process allows you to select which roles you want to cluster (e.g. file server, print server, virtual server), then sets up cluster resources and dependencies appropriately.

3. Improved SAN support
Windows server 2008 issues persistent reservations on shared storage to establish ownership of LUNs, it does not use bus resets any longer. Bus resets are disruptive on SANs where several systems on several platforms may share the same storage bus. This implies that the storage must support persistent reservations. Shared parallel SCSI is deprecated.

4. Changed quorum model
The administrator can choose the most appropriate quorum model for the configuration. Several are possible:

Majority node sets with witness disk: each node gets a vote and so does a witness disk. The cluster will survive the failure of any 1 vote (including the shared witness disk).
Majority of nodes: storage is replicated amongst them, but does not get a vote. The cluster is active until the majority of nodes is running.
Majority of nodes and file share witness: nodes get a vote and so does a file share on a separate server. This is ideal for geo-clusters, as the witness file share can be in a 3rd site. The geo-cluster will thus survive the loss of any one site.

5. Improved networking
The nodes need no longer be on the same private subnet and the timeout of the “ping” among nodes is configurable. This makes it possible to route private traffic between locations and removes any a-priori restrictions on distance. Obviously, practical restrictions remain and will depend on how much the clustered applications and their users will tolerate.

These are just a few of the innovations available in server 2008 clusters. You may want to try them out for yourself by building a simple cluster on a set of virtual machines. You don’t need shared storage any longer, but if you want to try a quorum with witness disk, you can set up one of those machines as an iscsi target.

Get started

If you want to try Windows Server 2008 clusters, a virtual lab is available on Technet Events.
There is also an excellent screencast by David Northey on http://edge.technet.com/Media/Windows-Server-2008-Failover-Clustering/
Official training will be launched shortly.

in reply to Windows Server 2008 Clusters

HPC Server 2008 webcast [HPC Server 2008 webcast]

gmarchetti — Wed, 23 Jan 2008 01:12:00 GMT

My introductory demo / webcast on HPC Server 2008 beta 1 has been recorded. It can now be viewed here.

I have received feedback from several people about it, requesting instructions to build the virtual machines I used and to repeat the demos. Hence, I have written a short document that will guide you through the set-up of a virtual HPC Server 2008 cluster on Hyper-V. It also provides you with some examples of what is possible with the new administration, diagnostic and scheduling tools. It is intended as a hands-on introduction to the salient features of HPC Server 2008 beta 1. It is not an official click-by-click manual and it must be considered work in progress. I will update it when beta 2 is released to cover new or changed features. No warranty is implied nor support promised. You try at your own risk ☺

The document is now available on https://windowshpc.net/Resources/Documents/2008b1_labs_and_demos.zip

in reply to HPC Server 2008 webcast

HPC and Gaming [HPC and Gaming]

gmarchetti — Tue, 15 Jan 2008 06:38:00 GMT

Who'd have thought that HPC would be applicable to videogames? Well, my mind was changed for me when we worked with CCP Games, makers of the largest online multi-player game, Eve. Here their developers explain why they are looking at high-performance computing and what is behind Eve. Enjoy :-)

in reply to HPC and Gaming

HPC in Action - 1 [HPC in Action - 1]

gmarchetti — Fri, 28 Dec 2007 21:48:00 GMT

Ever played with the "game of life"? Well, here's an example of the real thing: HPC used to study evolution and cellular dynamics. Bioinformatics is an interesting - if unusual - field of application. Computation and visualization go hand in hand, as one is required to make sense of the other. In this video you'll see examples of using WPF to display results computed by CCS v1 and v2. This is also one of the first examples of use of MPI.NET, a managed-code implementation of the message-passing libraries.

in reply to HPC in Action - 1

Linux / HPC Server 2008 dual-boot [Linux / HPC Server 2008 dual-boot]

gmarchetti — Fri, 14 Dec 2007 02:31:00 GMT

How to Build a Dual-Boot
Linux / Windows HPC Server 2008
System
Draft
Giovanni Marchetti, Technical Evangelist, Microsoft, gmarchet@microsoft.com
Overview
This document illustrates the steps necessary to build a dual-boot Linux / Windows HPC Server 2008 solution. Although it is intended for use on HPC clusters, the techniques illustrated are applicable with little modiﬁcation to deployment systems based on Windows Deployment Services, which is used in HPC Server 2008. The procedure can be summarized in:
1.Verify the existing Linux disk layout.
2.If necessary, modify the existing Linux partitions to free space for Windows installation.
3.Install Windows.
4.Set the active boot partition so that the desired o/s is started at the next reboot.

download the rest of this document from Windowshpc.net our HPC community site

in reply to Linux / HPC Server 2008 dual-boot

Windows HPC Server 2008 overview screencast, previously known as compute cluster v2 [Windows HPC Server 2008 overview screencast, previously known as compute cluster v2]

gmarchetti — Thu, 15 Nov 2007 08:00:00 GMT

This video will give you a quick tour of the 1^st beta of Windows HPC Server 2008, introducing the many advancements made in the area of deployment, management and scheduling. It is the first of a series of "hands-on" examples. If you've used compute cluster server or equivalent products, you'll like what we have in store for v2

Click here to watch the video, or right-click the download link below to save the video to your local machine.

in reply to Windows HPC Server 2008 overview screencast, previously known as compute cluster v2