The document introduces Oracle's Enterprise Cloud Infrastructure solution using SPARC T5 servers. It discusses Oracle's cloud strategy, challenges in building private clouds, and how Oracle addresses these challenges through optimized solutions like the Oracle Virtual Compute Appliance. It provides an overview of the Oracle Optimized Solution for Enterprise Cloud Infrastructure, including SPARC T5 servers, Oracle Solaris, Oracle VM Server for SPARC, and Sun ZFS Storage Appliance. Example configurations and best practices are also presented.
30. 30
Rack Layout
Half Rack Configuration
GbE Management Switch
10GbE ToR Switch #2
10GbE ToR Switch #1
SPARC T4-1 OC12 Server
7320 ZFSSA Storage Cluster Heads
ZFSSA Storage Shelf
SPARC T5-2 Virtualization node #1
SPARC T5-2 Virtualization node #2
SPARC T5-2 Virtualization node #3
SPARC T5-2 Virtualization node #4
SPARC T5-2 Virtualization node #5
• Standard HW Configuration
– 1 x SPARC T4-1 OC12 server
– 1 x 48 port Mgt Switch
– 1 x 7320 Storage
• 2 x Heads
• 1 x storage tray
– 5 x SPARC T5-2 virtualisation hosts
– 2 x 15KVA Single phase PDUs
31. 31
Rack Wiring
10GbE Wiring
QSFP to SFP+ cable ToR-Server wiring
• ToR #1
• Cable 1 – QSFP port 1
1. 7320 Head #1- port 1
2. 7320 Head #2- port 1
3. SPARC T4-1 OC12- port 1
4. - not used
• Cable 2 – QSFP port 2
1. SPARC T5-2 #1-1-1
2. SPARC T5-2 #1-2-1
3. SPARC T5-2 #2-1-1
4. SPARC T5-2 #2-2-1
• Cable 3 – QSFP port 3
1. SPARC T5-2 #3-1-1
2. SPARC T5-2 #3-2-1
3. SPARC T5-2 #4-1-1
4. SPARC T5-2 #4-2-1
• Cable 4 – QSFP port 4*
1. SPARC T5-2 #5-1-1
2. SPARC T5-2 #5-2-1
3. - not used
4. - not used
• ToR #2
• Cable 5 – QSFP port 1
1. 7320 Head #1- port 2
2. 7320 Head #2- port 2
3. SPARC T4-1 OC12- port 2
4. - not used
• Cable 6 – QSFP port 2
1. SPARC T5-2 #1-1-2
2. SPARC T5-2 #1-2-2
3. SPARC T5-2 #2-1-2
4. SPARC T5-2 #2-2-2
• Cable 7 – QSFP port 3
1. SPARC T5-2 #3-1-2
2. SPARC T5-2 #3-2-2
3. SPARC T5-2 #4-1-2
4. SPARC T5-2 #4-2-2
• Cable 8 – QSFP port 4*
1. SPARC T5-2 #5-1-2
2. SPARC T5-2 #5-2-2
3. - not used
4. - not used
KEY : SPARC T4-2 #1-1-1 = Server 1, PCI card 1, Port 1(top)
*Option to replace “QSFP to 4 x SFP+” Cables 4 & 8 with 4 x “SFP+ to SFP+” cables
Cable4
Cable3
Cable2
Cable1
Cable5
Cable6
Cable7
Cable8
32. 32
Rack Wiring
Management 1GbE Harness
1GbEManagementHarness
• ILOM connected via
NETMGT port
• 18 ports used on GbE
Management Switch
• ILOM NET MGT for
platform discovery and
management
• NET0 for OS provisioning
and monitoring
• Remaining 30 ports available
for other uses
Hello My name is … Today I would like to share with you Oracle’s comprehensive strategy for Cloud Computing. Today we will be focusing on Infrastructure as a Service today. We will share with you how Oracle can help as service providers and IT organizations find their own answers to their challenges.
Before we get started, I must share with you that information in this presentation is covered by this guidance.
IT focused on the wrong priorities! IT Ops orgs that have made significant progress with vmware are hitting a wall with respect to the next step for their cloud efforts. They over emphasized system virtualization, creating rampant vm sprawl which has in fact reversed the initial cost benefits of virtualization. They were only able to support simple and non-critical environments and hence their effort created no differentiation for the business. Despite having standardized on vmware, the virtualized infrastructure is too rigid to support value-added enterprise applications. The end result is that they cannot justify spending more on their virtualization projects. 09/10/13 10:20 AM EM11g @ Guggenheim
So way up at 10’ level - you can approach this in a number of ways between the two extremes of strategic and tactical. Cloud can really be a vehicle to implement strategic improvements to enterprise IT. This includes things like changing the cost structure of IT for lower TCO, it could a method to reduce complexity and enable rapid time to market, or even used to support other strategic initiatives, such as implementing a corporate green strategy. There are a lot of tactical motivators for cloud as well. A lot of these are also cost and ROI related, and can include things as simple as deploying virtualization technologies or things like enabling a more sophisticated way to pay for IT with new metering and chargeback capabilities. -----
We believe that enterprises are on a JOURNEY to cloud computing. Most will EVOLVE their current IT infrastructure to become more “cloud-like” – to become a better internal service provider to the lines of business, BUs, departments – to provide greater agility and responsiveness to business needs, higher quality of service in terms of latency & availability, and lower costs and higher utilization. This evolution will take time. Not only is the available technology evolving and advancing, but enterprises are also working on the new policies and processes needed. In many cases, the technical building blocks for cloud computing are available in advance of enterprise readiness, so we think that enterprises will evolve towards the right at different rates. The first step that many enterprises are taking is to move from a “Silo’ed” environment to a consolidated or virtualized environment. Most datacenters still have dedicated silo’s where each application runs on its own middleware, database, servers and storage. Each silo is sized for peak load, so there’s inherently a lot of excess capacity built in. Each silo is also different, leading to complexity and high costs to manage. Many organizations are moving from these silos to a virtual environment with shared services, dynamic provisioning and standardized configurations or appliances. This trend is very strong right now. Probably 80-90% of the companies I talk to are doing some form of consolidation right now, but they may be doing in only a portion of their datacenter, maybe 20-30%, so there’s more to do. <CLICK>From here, enterprises can evolve to a self-service private cloud with automated scaling (called policy-based resource management on the chart) and chargeback. Not every application benefits from self-service and elastic scalability, but some do, so enterprise are figuring that out and moving those first. Some organizations are not ready to implement full self-service, since that requires new policies and processes to be defined, and they may prefer allocation to pay-per-use chargeback models. There may be other challenges including gaining cross-organizational support, creating the business case and funding model, and various cultural issues. <CLICK>At the same time, Public Clouds are also evolving…various types of Service Providers and traditional ISVs are morphing to become SaaS, PaaS and IaaS providers. There are hundreds today, and adoption is proceeding at a rapid pace. But these providers are most often very specialized and isolated, and it’s difficult for customers to change providers. <CLICK>Ultimately, we think the evolution will move to “hybrid clouds” where a single application can span both private and public clouds and is managed in a federated manner. For this to happen, there need to be standards for interoperability and portability, and there needs to be technology to support such interoperability. We believe this hybrid model is the likely path for most of our customers. Each organization needs to create its own roadmap plan, and to decide what to move where and when. <CLICK> For many, the first step will be consolidation onto a shared platform and/or shared infrastructure. Then extending to a private PaaS or IaaS by adding self-service, policy-based scaling, metering/chargeback, etc. And use public clouds when the available offerings meet the requirements of the application.
This is a very high level picture of Oracle’s solutions for cloud. As you can see, Oracle offers rich solutions across private and public cloud for a broad range of uses unmatched by other solutions. On the left, you can see our products to help customers build and operate private clouds. On the right, you can see the Oracle Cloud which lets you use our public cloud services to accomplish the same objectives. Because we use the same technologies to power both private and public cloud, we make it easy for you to move your assets from one to the other. Let’s take a moment to look at our private cloud solution which is what Ed and most of you are going to use in order to power your private clouds.
Oracle private cloud solution is capable of powering your enterprise apps in a shared IT environment which encompasses your virtual as well as your physical resources. The richness of our solution for IaaS and PaaS offers more choices for your cloud deployment than any vendor. Enterprise capabilities to support your Fusion and Oracle Apps investment. Exadata and Exalogic have been tuned to support massive cloud environments running your Oracle Apps and databases. Management architecture is an integral aspect of the overall cloud architecture (unique) . The only business-driven application management, Complete cloud lifecycle management, and integrated cloud stack management are all unique capabilities which help you operate this cloud with ease, deliver the best customer experience and stay in control of the entire cloud.
When it comes to infrastructure, it’s important to understand Oracle’s overall strategy and how we’re investing in technologies. Customers can choose from Oracle’s comprehensive offerings the most suitable one for their environment. We also offer the customers the comfort of having the most suitable choices for deployment as well. We provide the best-in-class products and the engineered systems. Starting with Engineered Systems. This is where our customers see the most significant benefits for their cloud deployment. With integration from apps-disk , they reduce complexity and speed time to market And with the significantly increased performance , you can consolidate multiple apps on one system, essentially creating a cloud in a box. Oracle also offers hardware and software components that are best-of-breed by themselves. Customers can integrate these together or with 3 rd party components And then there’s the in-between approach , where you can take advantage of our “optimized solutions” which are best practice, optimization “recipes” or integrated configurations for lower risk and faster time to market. But since customers have a wide variety of needs, we go to even greater lengths with our Optimized Solutions, which provide flexible, predictable guided deployments that reduce risk, lower TCO and improve productivity that customer build on top of Best-in-Class products or the SPARC SuperCluster. With Oracle Optimized Solutions, you are able to select the best solution for a particular problem and then integrate the parts. In this case, you have a guided deployment, you know that all of the parts are designed to work together in an optimal fashion, you have the flexibility to change or substitute various components, but you know it’s going to work at the end and you purchased it all from one company so service is easier. A that’s a big win. Most Vendors and Customers are focused on Infrastructure Building Blocks as the future direction Incremental Improvements via low-level building blocks (e.g. Server virtualization): Most companies and vendors are able to achieve incremental improvements and efficiencies through server virtualization, and moving towards virtualized pools of compute, storage & network resources. Typically low-level hardware and hypervisor/server building blocks, not application oriented Limited to incremental improvements Oracle offers this too because there is real value in standardizing at the IaaS level
For each enterprise application environment, our engineering teams identify the best environment for running those applications – in some cases, that’s based on our Engineered Systems, in others we’re picking individual best-in-class server and storage components (SPARC, x86 or combinations). The engineering teams across the software and hardware products work together to identify, integrate and conduct full stack testing – resulting in documented best practices for deployment, tuning and getting the most out of your IT investments.
Here is the family of engineered systems and appliances. On the left you see the purpose build systems, with Exadata designed to run the Oracle database, and Exalogic designed to run Oracle middleware. As well as Exalytics and the database appliances. On the right is the general purpose SPARC SuperCluster which runs the Oracle parallel database, the Oracle fusion middleware parallel java system, and all your enterprise applications. The SPARC SuperCluster runs the Solaris operating system. And that’s extremely important, because it gives our SPARC customers a very, very smooth upgrade path. But it also gives our customers running older hardware from HP and IBM a powerful system to deploy Oracle applications for lower cost and with much better performance, and a management system for the entire stack. Exadata: OLTP/Data Warehouse, Database Consolidation, Database Upgrade Exalogic: Middleware Upgrade, Mainframe modernization, Middleware platforms standardization SPARC SuperCluster: general purpose, multi-tier applications, datacenter consolidation and virtualization
Main Talking Points: The word “Solutions” means many different things to many different people. Because we own the full stack, at Oracle, we base our solutions on optimization and testing done during the whole product development cycle. - We do real world workload testing on individual components as well as across the full stack. - Optimized Solutions are designed to take advantage of our extensive Hardware/Software Integration efforts. - Organizational Alignment is critical and enables faster innovation - --- Hardware and Software. Engineered to Work Together. http://www.oracle.com/us/corporate/features/engineered-173370.html Through high-performance systems engineering, Oracle delivers the best applications-to-disk environment by integrating its software, server, storage, and networking technologies to deliver some of the most scalable, reliable, and secure products available today. With the full software and hardware portfolio at its disposal, only Oracle designs and integrates everything from the semiconductor chip to the entire software stack to achieve performance levels that are unmatched in the industry. This tight integration between Oracle’s hardware and software improves system performance and provides better ways to access and manage information, all at a lower total cost of ownership. And because Oracle systems are built on open standards, they can be deployed in the datacenter as individual components, or as optimized solutions that are integrated from applications to disk. Integration is driven by Organizational Alignment Common Goals and Priorities are shared by the Teams Direct Collaboration of Engineering Teams enables Greater Ability to Share Core IP across products No NDA, No Attorneys, No Law Firms No Intermediate contacts, No Alliance Managers Organizational Alignment enables Faster Innovation and Integration Optimization across the Full Stack Optimization from Applications down to underlying Hardware We make sure that Software takes full advantage of unique features in Oracle Hardware (Oracle Flash Cache, on-board crypto, high thread count) while leveraging open standards and fully supporting other vendors Oracle is the one single vendor that can do all that Synchronization Hardware and Software Release Schedules No compatibility issues during deployment Patches and Updates will not break Customer Deployments
NOTE TO SPEAKER: Hammer out our the integrated , embedded features of SPARC T4 series servers throughout marked in BOLD . --------------- Oracle’s increased investments in SPARC servers with Oracle Solaris has resulted in a new processor design, the SPARC T4, which is at the heart of a new range of platforms. Oracle's SPARC T-Series servers running Oracle Solaris provide the most scalable, secure, and highly integrated platform for the optimized deployment of enterprise and mission-critical Oracle Enterprise applications as well as third party and custom applications. As a result, SPARC T4 servers achieve world record performance for a wide range of single and multi-threaded application workloads. They scale affordably by consolidating applications onto a single, highly expandable server with integrated , no-cost virtualization. You can also choose to employ Oracle Enterprise Manager Ops Center to improve the efficiency of IT operations with integrated life-cycle management, built-in automation, and flexible provisioning capabilities. Protect critical data and services with integrated wire-speed encryption using 16 industry standard ciphers and the highest levels of reliability, availability, and serviceability in its class. Other vendors offer similar functionality but usually this is provided at an added and frequently significant price on top of the initial hardware acquisition cost. Oracle’s SPARC T4 servers are the only products in the industry to offer two types of virtualization, on-chip cryptographic acceleration, and the operating system at no extra cost. Reduce data center expenses with built-in no-cost virtualization, management, security and networking.
Positioning: A scalable, intelligent set storage products running common software rich results in the economical management of large amounts of data. Ease of deployment, simplified management. This product line is able to address a wide spectrum of scalability, density, performance, and availability requirements. Workloads: Unstructured data, throughput oriented applications, mid-scale transaction processing, working data sets which fit in memory. Use Cases: File sharing, web storage, consolidation of virtualized environments requiring multiple data services, consolidation and backup/archival, streaming video/media, development environments
Oracle Solaris has been helping organizations evolve their datacenters toward a more service-oriented model by modernizing datacenter computing, networking, and storage management. Amongst the key innovations are built-in virtualization for the operating system with Oracle Solaris Zones (previously known as Containers), storage virtualization through the ZFS filesystem and the ability to safely analyze live production workloads using dynamic tracing (DTrace). With the advent of Oracle Solaris 11, significant new capabilities build on this baseline to improve traditional datacenter operations and pave the way to building enterprise-class clouds. Here the focus has been on enhancing and accelerating the ability to deliver services through extending virtualization capabilities to the network and the introduction of new installation and software management technologies. Oracle Solaris 11 is the first fully-virtualized OS in the industry delivering significant cost reductions through consolidation and software equivalents of expensive networking gear. Use Solaris 11 to start evolving your datacenter into a “Infrastructure as a Service” model to increase your ability to quickly deploy new services, respond to increased workloads and to efficiently provide heritage but necessary services.
Oracle offers a full portfolio of virtualization solutions to address your needs. SPARC is the leading platform to have the hard-partitioning capability that provides the physical isolation needed to run independent operating systems. Oracle VM Server for SPARC (previously called Sun Logical Domains) provides highly efficient, enterprise-class virtualization capabilities for Oracle's SPARC T-Series servers. You can create up to 128 virtual servers on one system to take advantage of the massive thread scale offered by SPARC T-Series servers and the Oracle Solaris. Many customers have already used Oracle Solaris Zones for application isolation. Solaris Resource Manager with fair-share scheduler, resource pools and processor sets are some key virtualization technologies in addition to Solaris Zones. Your organizations can couple Oracle Solaris Zones and Oracle VM Server for SPARC with the breakthrough space and energy savings afforded by Oracle’s SPARC T-Series servers to deliver a more agile, responsive, and low-cost environment. A very important “feature” of Oracle’s SPARC virtualization technologies is that they are no charge to the customer. No per core licenses are required as is the case with HP and IBM and their virtualization technologies. IBM and HP charge in the 10’s of thousands of $$$ for virtualization for their Itanium and Power servers. Oracle customers do need to pay for support if they want patches or upgrades to the SPARC virtualization technologies but no license fees. This means that customers can deploy the appropriate virtualization technology when they need to without having to worry about license fees. Some things to note: Solaris containers is available on all systems that run Solaris … even on non-SPARC systems.
Oracle VM Server for SPARC, previously called Sun Logical Domains, provides highly efficient, enterprise-class virtualization capabilities for Oracle’s SPARC T-Series servers. Oracle VM Server for SPARC leverages the built-in hypervisor to subdivide system resources (CPUs, memory, network, and storage) by creating partitions called logical (or virtual) domains. Each logical domain can run an independent operating system. Oracle VM Server for SPARC provides the flexibility to deploy multiple Oracle Solaris operating systems simultaneously on a single platform. This is the virtualization solution that fully optimizes Oracle Solaris and SPARC for your enterprise server workloads.
Solaris 11 offers both Zones for OS-level virtualization of workloads as well as OVM for SPARC and x86 hypervisor support. Over 4000 zero overhead zones can be used on a single server (128 VMs for OVM) to allow for high-density consolidation. System resources like IP stacks and ZFS datasets are easily assigned to each zone and VM if a strict multitenant model is required. Zones and VMs provide highly isolated application environments whose administration can be delegated to specific groups or customers for self-service IT management. Zones and VMs can be clustered to provide an even higher level of availability. Oracle VM for SPARC offers migration services for VMs.
8 Here's the architecture More details on the components that make up this Optimized Solution – which again just provides this ready to go virtualized infrastructure. It’s based on the Sun Blade systems which is a great foundation for IaaS - It supports the latest generation Intel Xeon processors (and SPARC blades) in a super dense, highly efficient chassis that has some really unique capabilities in how the network and I/O is set up to support shared service environments. And our ZFS Storage appliances with built in flash to build hybrid storage pools to optimized your storage architecture as well. (NOTE: There are a couple additional product slides in backup which you may use here if you want to cover more details on the hardware products) But suffice it to say, that this is a highly scalable, highly flexible hardware foundation that can be shipped as completely integrated, cabled up system that can be ordered pre-installed with Oracle VM and choice of both Oracle Linux and Oracle Solaris Operating Systems – and all managed together with Oracle Enterprise Manager 11g Ops Center - which has the very important job of managing both the physical and virtual infrastructure layers.
T5-2 has better single thread performance. T5-2 in general has better overall performance w/ 3.6GHz clock speed, higher memory link speeds, PCIe Bus bandwidth as well
I wanted to provide an example of a self-service cloud. This is actually Oracle itself. Over the last 6 or 7 years, we have built a self-service, private IaaS cloud for Dev/Test. Oracle has many developers working on many different products, and their need to access hardware for development and testing purposes fluctuates a great deal. Before, each product team had their own hardware, which was very expensive and incredibly inefficient. The decision was taken to centralize these servers and to share them across multiple teams. When a product is nearing a major new release, their need for computing capacity spikes, so the cloud can shift resources and priority to that team. When that team no longer needs capacity, it is re-allocated to other teams instead of sitting there idle. Developers also benefit from much more rapid access to computing resources. They have a self-service interface that allows them to submit jobs into a queue. They system prioritizes and assigns resources to the job, and automatically “cleans-up” for them afterwards. The developer simply gets their results back at a later point in time. You can see the benefits of this solution. Not only is this very cost effective and efficient, but it also accelerates product development and testing, which is actually a competitive advantage for Oracle.
Words from a partner … 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
In addition, there are a few other resources that you can use to optimize Oracle in your environment. Oracle Solution Centers offer help with production system requirements and design, prototyping , hands-on workshops that you can take advantage of when designing and deploying Oracle solutions. Oracle Optimized Solutions – In addition to the Optimized Solution for Oracle Database that we talked about today, you can find other Optimized Solutions on this website for deploying applications, middleware, and cloud. Oracle’s Systems Webcast Center is your single access point to dozens of webcasts on Oracle hardware systems and solutions. Get updates on new product features, IT best practices, uses for Oracle hardware in Oracle software environments, competitive comparisons, and much more. You can also join the Twitter conversation with hashtag #SPARC. Another great resource are Oracle Sales Consultants who have a wealth of knowledge working with Oracle’s software and hardware stack. If you’re ready to get started with an onsite workshop or assessment, contact your Oracle representative or authorized Oracle partner.