Friday Oct 04, 2013

Oracle x86 Systems: The Best x86 Platforms for Oracle Solaris

With adoption in over 50,000 businesses and institutions, Oracle Solaris has clearly established itself as the world leader in UNIX-based operating systems. Oracle Solaris includes many unique and innovative technologies that are not commonly available in other operating systems, such as Oracle Solaris ZFS, Dynamic Tracing (DTrace), predictive fault detection, built-in virtualization, and advanced security.

As a result of a broad strategic alliance between Oracle and Intel, Oracle Solaris has become widely deployed on Intel-based x86 systems from a variety of different hardware vendors. Intel has embraced Oracle Solaris as a mainstream UNIX operating system for enterprise-class, mission-critical systems based on Intel Xeon processors.

While Intel Xeon processor-based systems from any hardware OEM can be certified to run Oracle Solaris, Oracle has engineered Oracle Solaris to provide additional benefits when running on Oracle x86 systems. A primary example of this is the Oracle Solaris Fault Management Architecture (FMA). Developed and enhanced over the past decade, Oracle Solaris FMA is composed of a set of diagnosis engines that process raw error events from the hardware and provide an automated and intelligent method for problem diagnosis.

As a result of Oracle’s partnership with Intel, Oracle Solaris FMA was instrumented with diagnosis ability for DIMM and CPU correctible errors and fault detection for PCIe errors. Oracle Solaris also has the ability to offline individual processor strands, retire individual pages of memory and cease using problematic I/O devices.

While Oracle Solaris FMA is a built-in feature of the operating system, its capabilities can vary dramatically based on the server on which it is running, as is shown in the figure below:

Figure 1. Complete Oracle Solaris FMA capabilities are only available on Oracle x86 systems.

When running on Oracle x86 servers, Oracle Solaris automatically detects the presence of Oracle Integrated Lights Out Manager (Oracle ILOM), the embedded service processor that is part of every server from Oracle, and establishes a high speed connection between the two which allows both Oracle Solaris FMA and Oracle ILOM to have total visibility into the health of the entire system. Oracle’s x86 systems are the only Intel-based systems in which the service processor and Oracle Solaris are tightly integrated. As a result, organizations running Oracle’s x86 servers have the flexibility to choose how they want to monitor their servers, which simplifies their management infrastructure.

Oracle has also engineered Oracle Solaris FMA to provide special serviceability features when running on Oracle x86 systems. When a component fails in any of Oracle’s x86 servers, Oracle Solaris FMA will diagnose the problem and specify the failed component using the same nomenclature as the slot/component identification label silkscreened on the chassis. To further help in identifying and locating faulted components, when Oracle Solaris diagnoses a fault, a fault message is sent to Oracle ILOM, which in turn lights the appropriate fault indicators. Both of these allow for easier and faster service, enabling technicians to quickly restore services that were brought down by a fault.

Oracle Auto Service Request (ASR) is a secure, scalable, customer-configurable software solution for Oracle hardware products that resolves problems faster by providing automatic service request generation and priority service request handling for specific faults. If a fault is detected, the ASR software automatically creates a service request with the My Oracle Support portal. By automatically including the component part number in the service request to Oracle, Oracle ASR makes ordering replacement parts very simple. While non-Oracle hardware vendors have their own technologies for communicating problems back to their respective support organizations, those technologies are not tightly integrated with Oracle Solaris. Having a set of software and firmware components that have all been engineered and tested together is the only way to ensure accurate diagnosis of problems.

Oracle’s x86 systems are the best x86 platforms for Oracle software. Only Oracle provides customers with an optimized hardware and software stack that comes complete with choice of operating system, virtualization software and cloud management tools – all at no extra charge. Oracle’s optimized hardware and software stack has enabled a 10x performance gain in its engineered systems and has delivered world-record benchmark results. Oracle offers a wide range of Intel Xeon processor-based servers to suit the needs of enterprise class application that are required to conduct day-to-day operations:

 Product Name  Features  Primary Application
 Sun Blade X4-2B  Two-socket blade module based on Intel Xeon processor E5-2600 v2 product family  Compute node for clustered computing and virtualization needs in a blade form factor
 Sun Server X4-2 Two-socket 1RU server based on Intel Xeon processor E5-2600 v2 product family Compute node for clulstered computing and virtualization needs
 Sun Server X4-2L Two-socket 2RU server based on Intel Xeon processor E5-2600 v2 product family Storage node for large internal storage capacity needs
 Sun Server X2-4 Four-socket 3RU server based on Intel Xeon processor E7-4800 product family Consolidation node for efficient use of resources with high-performance and large memeory availability
 Sun Server X2-8 Eight-socket 5RU server based on Intel Xeon processor E7-8800 product family  SMP node for scale-up computing

Figure 2. Oracle x86 system portfolio

Oracle Solaris runs better and smarter on Oracle x86 servers. By engineering Oracle Solaris to work with Oracle’s x86 hardware, Oracle has engineered unique features that bring value to organizations that are looking to run their mission-critical applications. The advanced fault management and serviceability features, as well as improvements to installation and system management, make Oracle x86 systems the best hardware choice for running Oracle Solaris.

Pierre Lessard is a Principal Product Manager in Oracle’s Systems Group, focusing on x86 servers.  He spent the last fifteen years at Oracle in Operations and Engineering, working on both SPARC and x86-based systems.

Driving Database Innovation Down the Stack: Oracle Database and Oracle Linux on Oracle x86 Hardware

Oracle has always been a database company, but more recently we are driving database innovation even further down the stack through the operating system all the way down to the hardware.

Oracle x86 servers are the building blocks for Oracle’s engineered systems such as Exadata. These engineered systems run Oracle Database and Oracle Linux and are highly optimized to work together and provide unique features and benefits. For example, there is code in Oracle Linux and Oracle Database to make it run faster on Oracle x86, and there are firmware enhancements embedded in the server that make Oracle x86 the best x86 platform for running Oracle Database.

Oracle made these enhancements as part of the work to build engineered systems, but you can still take advantage of many of these features outside of engineered systems, if you are running Oracle Database and Oracle Linux on Oracle x86.

To understand how Oracle engineers its x86 servers to be the best for Oracle software, let’s first talk about Oracle’s overall design approach for building x86 systems. Oracle strives to make the most reliable, enterprise-grade servers because we know that this is what Oracle’s database customers want and expect.

Oracle x86 servers are built using Intel Xeon Processor family CPUs and chipsets, Intel NICs, and Intel SSDs. Oracle and Intel have a shared vision of creating the highest-performance, lowest power, and most reliable servers. Oracle collaborates with Intel to drive performance and reliability improvements at the component level to provide improved RAS features. Oracle also adds hundreds of firmware extensions to the storage subsystem and BIOS to ensure optimal database robustness.

In addition to the reliability built into each server, there is additional reliability that is engineered into Oracle Database when running on Oracle x86. This extra robustness comes from the process by which Oracle software is developed, tested, and deployed.

Oracle runs its own product development IT infrastructures on Oracle x86. In other words, software engineers developing Oracle Database write and unit test their code on Oracle x86, and Oracle Database builds are tested on a farm of Oracle x86 servers. Running Oracle Database and Oracle Linux on Oracle x86 is also a combination that Oracle uses to run its own cloud business, processing 3.2 billion business-critical database transactions per hour in Oracle’s Hosted cloud.

Through all of this additional testing and internal use of Oracle Database on Oracle x86, we uncovered many corner-case bugs in our storage subsystem, in our BIOS, and in other subsystems. And we worked through and fixed all of these bugs prior to general release. What this means is that customers running Oracle Database on Oracle x86, whether as part of an engineered system or a stand-alone server, are going to see fewer problems, less downtime, and reduced operating expenses.

Oracle and Intel not only collaborate closely to build servers but also to make sure that Oracle Database has maximum performance with Intel Xeon Processors. For example, Oracle and Intel worked together to accelerate encryption and decryption by three times by using AES-NI, improve the performance of checksum calculations in Oracle Database 12c by 40 percent, and add NUMA optimizations to improve locking by over 50 percent.

With the new Intel Xeon E5-2600 v2 series processors, Oracle will refresh its two-socket product line to take advantage of the improved performance, additional cores, and power savings. Oracle’s new Sun Server X4-2L server will support the fastest Intel Xeon E5-2600 v2 Processor family CPUs while also supporting over 50 TB of internal storage in only 2 rack units. And using Oracle’s Sun Flash Accelerator PCIe cards, Oracle Database I/O service times can be made fifteen times faster.

The net results from hundreds of firmware enhancements, hundreds more improvements to Oracle Linux, server components engineered for reliability and performance is a server that is designed for Oracle Database at all levels. And there are many more enhancements in the pipeline to further integrate hardware and software to provide a unified Red Stack.

Josh Rosen is a Principal Product Manager for Oracle’s x86 servers, focusing on Oracle’s operating systems and software. He previously spent more than a decade as a developer and architect of system management software. Josh has worked on system management for many of Oracle's hardware products ranging from the earliest blade systems to the latest Oracle x86 servers.

Monday Sep 30, 2013

Transform Data Center Performance with the Oracle SPARC Runner Video Game

Disaster has struck the Acme Co. data center! I’m Ora and I need your help! Aging, inefficient servers have left  the data center unable to respond to changing business needs. Tasked with increasing application performance, accelerating deployment speed, and consolidating multiple systems, Acme’s IT Manager called the best in the business—me, Ora, the SPARC Runner—to help him out.

But I can’t do this all by myself. I need your help navigating through five exciting, but treacherous data center levels. Help me jump, dash, and slide through complex and inefficient IT environments so I can:

  • Accelerate performance with the world’s fastest microprocessor
  • Increase speed with servers and OS that are co-engineered to do more together
  • Consolidate multiple legacy “IBS” servers onto a single SPARC architecture

See how many inferior "IBS" servers you can consolidate and challenge your friends to beat your score in this awesome new data center adventure video game, debuting at Oracle OpenWorld. Think you have what it takes? Play the new Oracle SPARC Runner and test your grit!

What do you think of the game and what was your high score?

Thursday Sep 19, 2013

Intel and Oracle: 21 Years Of Partnering to Deliver Innovation

As this year is heading to a close, it will complete 21 years of partnership between Intel and Oracle. For over two decades, both firms have worked together to drive increased performance for Oracle Database on Intel’s CPU architecture. In addition, Oracle and Intel have worked together to deliver innovation and hardened reliability for Oracle Solaris. However, what may get overlooked is Oracle’s commitment to the x86 architecture and Intel’s family of processors, through the continued development of Oracle's x86 servers

Fig-1: Oracle’s Solutions Built On The x86 Architecture & Intel’s CPUs

As mentioned above, one such example of this long standing partnership comes in the form of the advanced fault management capabilities of Oracle Solaris on Oracle’s x86 hardware. This enhanced system monitoring comes as a result of years of cumulative engineering investments made across multiple product families and generations. Combined innovations in Oracle Solaris FMA (Fault Management Architecture) and hardware design have resulted in a synergistic progression of new features that benefit end users. Oracle understands that organizations want a holistic approach to fault management in which all aspects of server health are examined together. This approach prevents system administrators from having to mine and collate error conditions from lots of different sources, such as operating system logs, device-specific tools, and service processors. Oracle Solaris FMA was instrumented with diagnosis ability for correctible memory errors for DIMMs, correctable errors (CEs) for CPUs, and fault detection for PCIe errors. For CEs, the component can be off-lined by Oracle Solaris FMA if the CEs occur too frequently. Also, Oracle Solaris FMA captures error state for CPU uncorrectable errors and reports them upon the next server reboot. Oracle’s x86 systems, all of which use Intel Xeon processors, were used to develop the integration between Oracle Solaris FMA and Intel Xeon processors, as well as the testing to ensure maximum visibility into each system. Lastly, these benefits are only available on Oracle x86 systems, as third party solutions cannot offer this level of integration between the operating system, service processor, and core computing components such as the CPU and memory.

Oracle’s ground-breaking Engineered Systems offer innovation through the integration of software and hardware to deliver unrivaled customer value. This disruptive design strategy, has created an entirely new market segment and system classification, and at the heart of these solutions are Oracle’s stand alone x86 servers. Oracle’s Intel Xeon based systems are the foundation for the majority of Oracle’s Engineered Systems including Exadata, Exalogic, Exalytics, Oracle Database Appliance, Oracle Big Data Appliance, Oracle ZFS Storage Appliance, and the newly introduced Oracle Virtual Compute Appliance. Oracle’s x86 servers offer the industry’s leading reliability and performance through hardening in these apex solutions, however they are also the computing backbone for Oracle’s cloud services and daily operations.  Oracle conducts 3.2 billion business critical database transactions per hour, across 20,000 Oracle x86 servers, as well as servicing over five million users through Oracle On Demand. This level of reliability is the reason so many customers use Oracle’s Engineered Systems and stand alone x86 servers for their mission critical or revenue generating IT resources.

Oracle’s hardware strategy will continue to be focused on enhancing the performance of Oracle software for greater business value to its customers. As Intel continues to refresh its enterprise class family of processors, customers can count on Oracle to follow the refresh schedule through deployment of Intel’s CPUs in Oracle’s stand alone servers, as well as its portfolio of x86 based Engineered Systems. On September 12, Oracle launched its three new systems based on the Intel E5-2600 v2 family of processors. The X4-2, X4-2L, and X4-2B are the newest two-socket systems from Oracle, and are designed for clustered databases, virtualized cloud applications and mission critical business services.

Look for additional blogs to come, detailing how Oracle offers greater database performance when running on Oracle Linux and Oracle x86 systems, and a deeper dive into the additional features of Oracle Solaris on Oracle x86 servers. If you are attending Oracle OpenWorld later this month, be sure to attend Intel and Oracle’s joint sessions covering these topics.

Oracle Open World x86 Sessions:
The Oracle x86 Advantage: An Overview of Oracle’s x86 Products and Value Proposition
Westin San Francisco – Concordia Room, Monday, 9/23, 10:30 a.m.-12:00 p.m.

Driving Database Innovation: Oracle Database and Oracle Linux on Oracle x86 Hardware
Westin San Francisco – Olympic Room, Monday, 9/23, 3:00-4:30 p.m.

Edgar Haren is a Principal Product Director at Oracle for Oracle’s x86 systems and Oracle’s Virtual Compute Appliance. He previously spent several years in product development and marketing focused on High Performance Computing. Edgar has 8 provisional patents and 11 public disclosures focusing on server power solutions, mobile power, wireless audio, direct audio, USB switching, physical security, video projection and ergonomics.

Tuesday Sep 17, 2013

Oracle OpenWorld Preview Guide: Servers and Engineered Systems

Oracle OpenWorld

For the hardware fanatics out there, Oracle OpenWorld is just about here and we have all the details you need to get the most out of your time. The keynote will be given by Oracle CEO Larry Ellison, on Sunday, September 22 at 5:00 p.m. to kick off the whole event. This will be held in Hall D at Moscone North. Don’t miss it!

Here are some of the Systems highlights during the 5-day event:

  • Over 70 Server, Engineered Systems and Oracle Solaris sessions plus many others covering database and systems management, virtuaization, security and more
  • Meet with Oracle Server developers and product teams
  • 50+ demos running on Oracle Servers and Engineered Systems

Many of the System sessions will be held in the Westin San Francisco on Market Street.

Some highlights of the upcoming sessions include:

There will also be sessions specifically for SPARC servers, Engineered Systems and Solaris. You can learn about Using Solaris Zones to Enhance Cloud Platforms, Breakthroughs in SPARC Processor Technology and In-Memory Computing and What’s New and What’s Coming in Oracle Exadata.

After your sessions, you will be given the opportunity to meet with Oracle developers and product teams. Lunch, coffee, and recharging stations will be provided.

Also, you will be given the chance to try things out in our hands-on lab sessions, including:

  • Servers: Database Cloning, Oracle FLEXCUBE, In-Memory Features of M-Series, Consolidation with Oracle Database 12c, Oracle x86 System Advantages, Improving SAP Operations and Performance, Virtualization on SPARC, and many more
  • Engineered Systems: SuperCluster T5, Exadata, In-Memory DB for Exalytics, Oracle Database Appliance, Fusion Middleware on Exalogic, Oracle Virtual Compute Appliance, and check out the Engineered Systems Showcase outside keynote Hall D
  • Solaris: Optimizing Oracle Databases, Extreme Availability, Virtualization and Consolidation, Software Defined Networking

Challenge yourself to soak up as much information as you can! Oracle OpenWorld is only one week long! Don’t forget about Customer Appreciation Night on Wednesday, September 25. It’s a chance to let loose, soak up a great view of San Francisco, and dance the night away with Maroon 5 and The Black Keys. It’s the best party in town!

Follow this link for the full schedule of events:

Which Systems event are you looking forward to at Oracle OpenWorld?

Thursday May 23, 2013

“As The World 'Goes Software,' Oracle Has An Upside In Hardware” says ESG Analyst

Last month, Oracle invited industry analysts for an annual deep dive on what’s to come. One of these analysts – ESG’s senior analyst Mark Peters – recently posted an update of how he spent some of his time at the event. We totally recommend you visit his blog and check out the post, but here are a few highlights:

Diverse Crowd
Peters was struck by the fact that he wasn’t seeing a lot of familiar faces in the crowd. He later realized the reason for that was because Oracle doesn’t just have a hardware message – it also speaks to the software side of the equation. 

“The opportunities for Oracle to make significant progress in terms of market share and revenue lie in markets that it does not yet dominate. It’s a crucial point for this growing systems house.”

That’s a crucial point we’re well positioned for. Our complete stack of offerings provide world-record-breaking performance by fusing critical pieces of software and hardware in ways that deliver cost-performance unlike anything the tech industry has seen.

Bullish on Storage
Mark filmed a brief interview with Oracle’s Phil Bullinger, senior vice president of storage. Mark and Phil go into great details on the ZFS Storage business.

Magnetic Personality
Mark also interviewed Jim Cates, vice president of hardware development, regarding what’s coming for tape storage from Oracle – and more.


Thursday May 02, 2013

New Storage Podcast Series Highlights 10 Important Purchase Considerations

Whether you are new to storage or a seasoned professional, deciding which storage systems to buy can have a big impact on how your business performs. There are a lot of factors to consider and each one should be weighed in light of the impact it will have on your applications, business processes, risk exposure, management overhead and of course your budget. I have started a series of podcasts to examine the top 10 considerations you should keep in mind when deciding which storage system you should go with as well as the impact those considerations might have for your business. Luckily, we have many great people at Oracle with deep storage experience, so I rounded up three individuals who can really help examine common issues like performance, scalability and high availablity. I will be releasing roughly one episode a week so check back often. You can listen right from the webpage below or just subscribe with your favorite podcatching app.

Podcast: Making The Right Storage Choices - A Ten Part Series

Wednesday Apr 17, 2013

Oracle Shines at NAB Show

I attended the 2013 NAB (National Association of Broadcasters) show in Las Vegas last week and, wow, I have to say I was a bit blown away.  Of course, it would be prudent to say that it was my first time attending NAB, so that probably had a lot to do with my overall impression.

NAB is one of the largest trade shows that comes to Las Vegas each year and the pure size of the show is impressive.  The show is a who's who of technology suppliers to the media and entertainment (M&E) industry.  From the usual tech industry suspects including Oracle, IBM, HP, Microsoft, Cisco, EMC, Adobe, etc. to M&E specific suppliers such as Grass Valley, Front Porch Digital, Black Magic, AJA and Avid Technology, anyone who is anyone, or desires to be someone, in the M&E space was there with all their latest software, hardware and other gadgets and gee whiz gizmos.  Everything you need for digital content creation, post production work, workflow management, content distribution and content delivery systems.  

The amazing thing for me to witness was the complete transformation of the industry to the digital age. The transformation from analog to digital probably began in earnest in the M&E industry about fifteen years ago and it is evident that the transformation is now nearly complete.  The fact that the above mentioned, traditional IT suppliers were well represented with large booths is a good indication of the progress of the M&E industry to an all digital workflow... from content creation to content delivery and finally to digital archiving.

You could consider the NAB show as an annual barometer for how the traditional IT community is embracing the media and entertainment industry as a truly legitimate, high growth opportunity.  As far as vertical market opportunities go these days, you hear a lot about the healthcare, energy, retail and financial industries as the top opportunities, but from the perspective of the floor of the NAB show, you would have to put the M&E industry right at the top with those others.  As the M&E industry completes it's transition to digital, the infrastructure required to drive it becomes a multi-billion dollar opportunity for both software and the underlying hardware infrastructure which includes plenty of high powered servers, storage and networking.  That's why Oracle was at the show with an impressive booth full of M&E solutions including:

Traffic in the Oracle booth was solid for the entire four days of the show with many industry leaders stopping by for in-booth demonstrations and meetings including HBO, Turner Broadcasting, Front Porch Digital and Harris to name just a few.  Many of these discussions revolved around requirements for moving digital media assets through the workflow process, as well as the integration of digital archiving throughout that workflow.  Many customers were interested in understanding how the new LTFS specification for digital tape has emerged as the  enabler for simplifying the integration a digital repository or archive into the DAM/MAM (digital asset management/media asset management) workflow. As a co-chair of the SNIA (Storage Networking Industry Association) committee responsible for the LTFS specification, Oracle is uniquely qualified to guide M&E customers through that discussion. 

And as it is doing with LTFS, Oracle will be there with the right technology and the right solutions to assist the media and entertainment industry as it continues it's digital media revolution.  So don't forget to look for us at the 2013 IBC show on September 13-17 in Amsterdam and next year's NAB show in Las Vegas on April 7-10, 2014.

Monday Apr 08, 2013

Power DBaaS with Sun ZFS Storage Appliance

If you have been thinking about setting up a DBaaS (Database as a Service) for your company you won't want to miss this webcast about implementing DBaaS using Oracle Enterprise Manager 12c and Oracle Sun ZFS Storage Appliance. In this webcast you will learn about the features and benefits that the Sun ZFS Storage Appliances provide when used with Enterprise Manager 12c DBaaS Snap Clone and watch a live demo of this exciting new feature. Snap Clone leverages Oracle's own Sun ZFS Storage Appliance system to clone even the largest database in minutes while at the same time providing unmatched performance and efficiency. Users can create their own copies for functional testing and development without consuming large amounts of disk space.

Webcast: Database Provisioning Made Easy - Using Enterprise Manager 12c DBaaS Snap Clone and Sun ZFS Storage Appliance
April 11 - 10:00 a.m. PT

Register Now!

Wednesday Apr 03, 2013

Unleash the Full Power of Oracle's new SPARC T5 and M5 Servers with Sun ZFS Storage Appliance

FDR wasn’t talking to storage architects when he famously said “We have nothing to fear but fear itself,” but with Oracle’s launch of powerful new SPARC servers, he could have been.  Just think, the SPARC T5-8 server packs up to 1,024 threads and 4TB of memory into 8U of rack space – a capability that up until a couple of years ago was reserved for the world’s fastest supercomputers. As a result, storage architects are concerned that their legacy NAS and SAN systems won’t be able to keep up with these new computational powerhouses. Is their fear justified?

The answer is an unequivocal “Yes, “but they can do something about it.

Over the years I’ve seen lots of companies stick with existing storage when they move to new, more powerful servers.  I understand the reasoning – they’re just running the same applications and databases, so they don’t need new storage.  But their users won’t use the applications in the same way  as before even if that was the plan.  For example, business intelligence apps will run faster, so end-users want to analyze larger problems or run more predictive models,  putting unforeseen pressure on the storage systems, which can easily lead to fear. 

To assuage your fear, you need to ensure that your new SPARC T5 servers are complemented by equally powerful storage systems that can keep up with the I/O loads the servers generate and yet be agile enough to change when application requirements dictate. The good news is that Oracle has a storage solution that meets these needs.

Oracle’s Sun ZFS Storage Appliance are specifically engineered to support the new SPARC servers with more compute threads to serve more concurrent I/O requests, more memory and Flash caches to efficiently serve massive numbers of VMs, and more memory bandwidth to move data through the system as it is needed.

And, it doesn’t stop there. 

Many of the SPARC T5 servers that we ship will run Oracle Database, and Oracle’s efforts to engineer hardware and software together extend beyond the database and the server to include the three-way optimization of Oracle software, Oracle servers, and Oracle storage.  

It only makes sense that Oracle would be able to engineer the best storage for Oracle Database and SPARC T5 systems.  There are many reasons why this is so, but three of the ones we highlight in our press release are:

1. Oracle Database benefits from high levels of parallelization in both SPARC T5 server and Sun ZFS Storage Appliances. 

2. Sun ZFS Storage Appliances are the only NAS systems that support Oracle Solaris Cluster and work with SPARC T5 systems to deliver sub-second application failover for mission-critical applications that experience a server or software problem. 

3. Sun ZFS Storage Appliances deliver multi-faceted I/O stack optimization for Oracle Database that that increase performance and simplify the creation and development of both production and test and development environments. 

The bottom line is that you can update your compute infrastructure independently of your storage and risk the consequences, or you can incorporate Sun ZFS Storage Appliances and get the only storage specifically engineered to optimize SPARC T5 environments.

To find out more about the Sun ZFS Storage Appliance and how it supports Oracle Database environments download the Transforming Business with Oracle Optimized Storage for Oracle Database 11g white paper and see how Oracle IT Relies on Sun ZFS Storage Appliance to power the Oracle Cloud.

Monday Mar 04, 2013

Oracle x86 Systems – Putting the Infrastructure in IaaS (Part One of Three)

Recent IDC cloud research shows that worldwide revenue from public IT cloud services will reach $72.9 billion in 2015, representing a compound annual growth rate (CAGR) of 27.6%. Specifically, Infrastructure as a Service (IaaS) will have shown a 33% CAGR dating from 2011-2014. There are various forces driving this meaningful growth, and different sources all agree that the SMB-Cloud Market will likely grow to over $50B by 2014, with the Enterprise space reaching $20B. Small- and medium-sized businesses are migrating to the cloud to reduce the operating and capital expenditures associated with hardware and software, as well as costs associated with support and maintenance.  Enterprise customers are also taking advantage of the cloud, but in a different light. Enterprise customers are segmenting their IT into portions that generate revenue and portions that are considered standard IT, and capable of being moved to the cloud. The common theme is that companies want to simplify their IT and focus their internal resources on areas that allow them to meet the expectations of their internal and external customers, as well as keep pace with their competition.

Oracle understands these needs and offers a complete ecosystem of solutions jointly designed with their x86 portfolio of systems to offer its customers meaningful innovation centered on delivering business value and the most efficient path to the cloud. Oracle’s IaaS solution includes their x86 standalone servers, the innovative Oracle ZFS Storage Appliance, the choice of operating systems in Oracle Linux or Solaris, virtualization software with Oracle VM, Oracle Enterprise Manager 12c for complete system management, and the newly introduced Oracle Virtual Networking, which combines hardware and software for tremendous reductions in Total Cost of Ownership (TCO).

Figure-1. Oracle Optimized Solution for Enterprise Cloud Infrastructure

Oracle x86 Red Stack 

Oracle’s IaaS x86 Block has been implemented in conjunction with Oracle’s middleware and applications by several cloud solution providers, including dcVAST, an IT and Managed Services company serving customers ranging from middle market to Fortune 500 organizations. Prior to 2011, dcVAST has a mixed solution consisting of Oracle software, Red Hat Linux (RHEL), along with HP and Dell servers virtualized using VMware. However, they encountered several challenges including a lack of service flexibility, limited scalability, ineffective solution support, and rising costs associated with VMware and RHEL. This complexity led David Cintron, vice president of Managed Services at dcVAST, to migrate their IaaS services to Oracle’s enterprise cloud infrastructure architecture, which provides a single-vendor solution for the entire hardware and software stack, and is designed for rapid deployment. The solution dcVAST implemented to replace the complex and mixed vendor installation consisted of Oracle’s x86 blade systems, Oracle Linux, Oracle VM, Oracle ZFS Storage Appliance, Oracle’s Fabric Network and Oracle Enterprise Manager Ops Center 12c for ubiquitous management.  Cintron points out, “The fact that Oracle VM and the Hypervisor are included means there is less cost to our customers. Also, our engineering staff finds the Oracle Enterprise Manager Ops Center, which includes a patch management tool, especially useful. Without it, we would have had to buy another product or cascade into operating system management around Red Hat. We need to run a supported Linux environment and be able to rely on security and patch fixes, and now this is part of the solution with Oracle hardware. This is a valuable added benefit for us and our ability to service our customers.”

Oracle Enterprise Manager Stack for x86 Systems

Six months after implementation, the Oracle solution is on track in delivering 50 percent TCO savings based on the low acquisition cost and ongoing operational efficiencies. Only with Oracle’s x86 systems with a support contract, do customers receive their choice of Oracle Linux or Solaris, Oracle VM, and Oracle Enterprise Manager Ops Center 12c at no additional cost, offering valuable savings for customers to reinvest in revenue generating activities. Don Swanson, CEO of dcVAST, explained, "The Oracle Optimized Solution for Enterprise Cloud Infrastructure provides a customizable solution that can be tailored specifically for our clients running Oracle systems. It delivers smoother deployment and increased functionality of Oracle ATG Commerce and cloud services. If customers are running Oracle software and they are moving to the cloud, this is the solution they should be using because true to form, Oracle runs better on Oracle.”

Now that you have been introduced to Oracle’s IaaS architecture for public or private clouds, look for part two and three of this topic, where I will cover deeper insights into the innovations found within each architectural element, and how they integrate to provide a differentiated solution, aimed at helping customers meet the ever growing demands on their cloud implementations.

Edgar Haren is a Principal Product Director at Oracle for Oracle’s x86 systems and solutions. He previously spent several years in product development and marketing focused on High Performance Computing. Edgar has 19 provisional patents and public disclosures focusing on server power solutions, mobile power, wireless audio, direct audio, USB switching, physical security, video projection and ergonomics.

Tuesday Jan 15, 2013

Oracle IaaS with Capacity on Demand – Engineered Systems Hardware for a Monthly Fee. Possibly the Greatest Thing Since Sliced Bread.

It’s no mystery that Oracle offers the most comprehensive and flexible portfolio of cloud products and services in the industry. From public cloud to private cloud, our strategy is to empower and support customers to make the best decisions with products and services that address their needs throughout their journey to the cloud. Today, we’re excited to announce the addition of Oracle Infrastructure as a Service with Capacity on Demand (Oracle IaaS).

Engineered SystemsWhat is Oracle IaaS you say? It’s Oracle’s newest private cloud offering that enables organizations to deploy Oracle Engineered Systems, including Oracle Exadata Database Machine, Oracle Exalogic Elastic CloudOracle Exalytics In-Memory Machine, Sun ZFS Storage Appliance,and Oracle SPARC SuperCluster, in their own data centers, behind their firewall – all for a monthly fee.

Yes, you heard that right. Now you can have Oracle Engineered Systems hardware deployed on-premise without upfront capital expenditures in a pure OpEx model in which Oracle still owns the hardware. With Oracle IaaS, instead of buying or leasing the hardware, you can now spread the cost over time by paying a simple monthly fee – there’s no financing or complex lease terms involved.

What’s Capacity on Demand? Well for customers who want the power, security, and reliability of an on-premise engineered system, but don’t necessarily need all of that computing capacity all the time, each Oracle IaaS system includes additional CPU capacity that can be enabled at any time to handle peak workloads and disabled when no longer needed. Capacity on Demand helps keep infrastructure costs down because customers only pay for the additional CPU capacity during the months it’s used, while providing extra juice whenever workloads spike or business requirements change.

Wait, there’s more. Oracle Infrastructure as a Service includes industry-leading services that maximize performance, reliability and security. With the monthly fee, customers also receive Oracle Premier Support for Systems, Oracle Platinum Services, and the new Oracle PlatinumPlus Services which provide quarterly proactive analysis and advisory services to ensure that customers’ systems are configured optimally – all at no extra cost.

Don't just take my word for it. Learn more about Oracle IaaS with Capacity on Demand at Also, join Mark Hurd and Juan Loaiza on January 22nd for the LIVE webcast Introducing Oracle Infrastructure as a Service.

Tuesday Nov 20, 2012

SPARC at 25: Past, Present and Future

Join us online to celebrate a quarter-century of innovation. Watch Scott McNealy, Bill Joy, and Andy Bechtolsheim along with other significant SPARC contributors discuss the challenges and rewards of consistently redefining the limits of enterprise IT. Hear Mark Hurd and John Fowler talk about the aggressive plans for SPARC’s future. All of this was recently captured in video at the SPARC anniversary event held at the Computer History Museum in Mountain View, California.

SPARC Anniversary

In addition to getting unique insights from the people behind 25 years of SPARC technology, you can access exclusive content and resources, read case studies and e-Books, view webcasts and infographics, and more.

Be sure to take some time to rediscover why and how SPARC was developed, the considerable impact it had on the entire IT industry, and the continuing innovations coming in the future.

Thursday Nov 15, 2012

The SPARC SuperCluster

Oracle has been providing a lead in the Engineered Systems business for quite a while now, in accordance with the motto "Hardware and Software Engineered to Work Together." Indeed it is hard to find a better definition of these systems. 

Allow me to summarize the idea. It is: 

  • Build a compute platform optimized to run your technologies
  • Develop application aware, intelligently caching storage components
  • Take an impressively fast network technology interconnecting it with the compute nodes
  • Tune the application to scale with the nodes to yet unseen performance
  • Reduce the amount of data moving via compression
  • Provide this all in a pre-integrated single product with a single-pane management interface

All these ideas have been around in IT for quite some time now. The real Oracle advantage is adding the last one to put these all together. Oracle has built quite a portfolio of Engineered Systems, to run its technologies - and run those like they never ran before. In this post I'll focus on one of them that serves as a consolidation demigod, a general purpose engineered system. 

As you probably have guessed, I am talking about the SPARC SuperCluster. It has many great features inherited from its predecessors, and it adds several new ones. Allow me to pick out and elaborate about some of the most interesting ones from a technological point of view. 

The SPARC SuperCluster

I. It is the SPARC SuperCluster T4-4. That is, as compute nodes, it includes SPARC T4-4 servers that we learned to appreciate and respect for their features:

  • The SPARC T4 CPUs:
    • Each CPU has 8 cores, each core runs 8 threads. The SPARC T4-4 servers have 4 sockets. That is, a single compute node can in parallel, simultaneouslyexecute 256 threads. Now, a full-rack SPARC SuperCluster has 4 of these servers on board. Remember the keyword demigod. 
    • While retaining the forerunner SPARC T3's exceptional throughput, the SPARC T4 CPUs raise the bar with single performance too - a humble 5x better one than their ancestors. 
      • actually, the SPARC T4 CPU cores run in both single-threaded and multi-threaded mode, and switch between these two on-the-fly, fulfilling not only single-threaded OR multi-threaded applications' needs, but even mixed requirements (like in database workloads!).
    • Data security, anyone? Every SPARC T4 CPU core has a built-in encryption engine, that is, encryption algorithms cast into silicon
    • A PCI controller right on the chip for customers who need I/O performance. 
  • Built-in, no-cost Virtualization: 
    • Oracle VM for SPARC (the former LDoms or Logical Domains) is not a server-emulation virtualization technology but rather a serverpartitioning one, the hypervisor runs in the server firmware, and all the VMs' HW resources (I/O, CPU, memory) are accessed natively, without performance overhead. 
    • This enables customers to run a number of Solaris 10 and Solaris 11 VMs separated, independent of each other within a physical server

II. For Database performance, it includes Exadata Storage Cells - one of the main reasons why the Exadata Database Machine performs at diabolic speed. What makes them important?

  • They provide DB backend storage for your Oracle Databases to run on the SPARC SuperCluster, that is what they are built and tuned for DB performance. 
  • These storage cells are SQL-aware
    • That is, if a SPARC T4 database compute node executes a query, it doesn't simply request tons of raw datablocks from the storage, filters the received data, and throws away most of it where the statement doesn't apply, but provides the SQL query to the storage node too. The storage cell software speaks SQL, that is, it is able to prefilter and through that transfer only the relevant data. With this, the traffic between database nodes and storage cells is reduced immensely. Less I/O is a good thing - as they say, all the CPUs of the world do one thing just as fast as any other - and that is waiting for I/O. 
    • They don't only pre-filter, but also provide data preprocessing features - e.g. if a DB-node requests an aggregate of data, they can calculate it, and handover only the results, not the whole set. Again, less data to transfer. 
  • They support the magical HCC, (Hybrid Columnar Compression). That is, data can be stored in a precompressed form on the storage. Less data to transfer. 
  • Of course one can't simply rely on disks for performance, there is Flash Storage included there for caching. 

III. The low latency, high-speed backbone network: InfiniBand, that interconnects all the members with:

  • Real High Speed: 40 Gbit/s. Full Duplex, of course. Oh, and a really low latency. 
  • RDMA. Remote Direct Memory Access. This technology allows the DB nodes to do exactly that. Remotely, directly placing SQL commands into the Memory of the storage cells. Dodging all the network-stack bottlenecks, avoiding overhead, placing requests directly into the process queue. 
  • You can also run IP over InfiniBand if you please - that's the way the compute nodes can communicate with each other. 

IV. Including a general-purpose storage too: the ZFSSA, which is a unified storage, providing NAS and SAN access too, with the following features: 

  • NFS over RDMA over InfiniBand. Nothing is faster network-filesystem-wise. 
  • All the ZFS features onboard, hybrid storage pools, compression, deduplication, snapshot, replication, NFS and CIFS shares
  • Storageheads in a HA-Cluster configuration providing availability of the data 
  • DTrace Live Analytics in a web-based Administration UI
  • Being a general purpose application data storage for your non-database applications running on the SPARC SuperCluster over whichever protocol they prefer, easily replicating, snapshotting, cloning data for them. 

There's a lot of great technology included in Oracle's SPARC SuperCluster, we have talked its interior through. As for external scalability: you can start with a half- of full- rack SPARC SuperCluster, and scale out to several racks - that is, stacking not separate full-rack SPARC SuperClusters, but extending always one large instance of the size of several full-racks. Yes, over InfiniBand network. Add racks as you grow. 

What technologies shall run on it? SPARC SuperCluster is a general purpose scaleout consolidation/cloud environment. You can run Oracle Databases with RAC scaling, or Oracle Weblogic (end enjoy the SPARC T4's advantages to run Java). Remember, Oracle technologies have been integrated with the Oracle Engineered Systems - this is the Oracle on Oracle advantage. But you can run other software environments such as SAP if you please too. Run any application that runs on Oracle Solaris 10 or Solaris 11. Separate them in Virtual Machines, or even Oracle Solaris Zones, monitor and manage those from a central UI.

Here the key takeaways once again:

The SPARC SuperCluster:

  • Is a pre-integrated Engineered System
  • Contains SPARC T4-4 servers with built-in virtualization, cryptography, dynamic threading
  • Contains the Exadata storage cells that intelligently offload the burden of the DB-nodes 
  • Contains a highly available ZFS Storage Appliance, that provides SAN/NAS storage in a unified way
  • Combines all these elements over a high-speed, low-latency backbone network implemented with InfiniBand
  • Can grow from a single half-rack to several full-rack size
  • Supports the consolidation of hundreds of applications

To summarize: All these technologies are great by themselves, but the real value is like in every other Oracle Engineered System: Integration. All these technologies are tuned to perform together. Together they are way more than the sum of all - and a careful and actually very time consuming integration process is necessary to orchestrate all these for performance. The SPARC SuperCluster's goal is to enable infrastructure operations and offer a pre-integrated solution that can be architected and delivered in hours instead of months of evaluations and tests. The tedious and most importantly time and resource consuming part of the work - testing and evaluating - has been done. 

Now go, provide services.  

-- charlie  

[Read More]

Oracle engineers hardware and software to work together in the cloud and in your data center. For more information about Oracle (NASDAQ:ORCL), visit


« April 2014