What Is Hyperconvergence? - SLIDEBLAST.COM

PUBLIC SECTOR

LENOVO INTERNAL AND PARTNER USE ONLY

What Is Hyperconvergence? According to IDC, “hyperconverged infrastructure essentially collapses core storage, computing, and networking functions into a single software solution or appliance. Hence it is simply a more tightly integrated converged system with compute, storage, and networks decoupled from the underlying infrastructure and defined/ configured at a software level.”1

...hyperconverged infrastructure essentially collapses core storage, computing, and networking functions into a single software solution or appliance.

Intel Inside®. Powerful Solution Outside. Intel® Xeon® Processors.

There is more than one way to deliver a hyperconverged solution: as a hardware-software appliance in a pre-defined configuration, or as a flexible solution based on a purely software-defined architecture. In either case, the principal characteristic common to either approach is that expensive, monolithic storage frames and the physical SAN are both eliminated. The result is a unified compute and storage infrastructure that is simpler to manage, reduces capital and operational expenses, and doesn’t require vendor-specific SAN or storage expertise.

WHY HYPERCONVERGE? A full 44 zettabytes: That’s how much data there will be on the planet by 2020.2 It’s 10 times more data than there was in 2014 — and far more data than traditional SAN-based IT infrastructures can ever hope to process, store, and analyze. It is not just the volume of data, but the rate at which it is created. IDC is reporting that 90 percent of all the digital data stored today was created in the last few years.3

Predicting the need for storage capacity becomes a struggle and a challenge that needs a systematic and predictable resolution. Hyperconvergence replaces the proprietary hardware-defined storage and physically converged infrastructure with a softwaredefined storage infrastructure that is virtually converged within the hypervisor tier, hence becoming “hyperconverged.” In hyperconvergence, hardware-defined elements for computing, storage, networking, and management converge into a highly simplified, software-defined virtual architecture. The result is a dramatically simplified data center infrastructure that transcends the limitations of traditional systems. It improves scalability and performance, enhances manageability, and dramatically reduces the cost of operations in the long-term.

THE EVOLUTION OF THE IT INFRASTRUCTURE As organizations and institutions continue to look for ways to handle growing demands on the technology infrastructure, the limitations of traditional hardware-defined infrastructures create pain points that cannot be resolved with traditional thinking. With SDS moving into the servers, new thinking and new architectural approaches are possible. The fusion of SDS and server virtualization enables hyperconvergence, moving Lenovo® customers a giant leap toward a 100 percent software-defined future.

PUBLIC SECTOR


EVOLUTION OF THE IT INFRASTRUCTURE

EASIER TO MANAGE Software-defined infrastructures are easier to manage than hardware-defined systems. Because they eliminate the need for multiple hardware systems for server, storage, networking, and other capabilities, there’s no need to juggle multiple management tools and interfaces. Users can now manage all resources from a common management framework. SEAMLESS TO SCALE Hyperconverged systems scale seamlessly and dynamically. Scaling to meet increasing demand is a matter of simply adding nodes for front-end processing or back-end capacity any time more compute power or storage is needed — or just as easily removing capacity for either when demand goes down. This elastic-scale approach overcomes the complexity of scaling traditional hardware-defined systems. In addition, hyperconverged infrastructure improves infrastructure ROI with:

Hyperconvergence is the next natural step in the evolution of the data center. Complex IT infrastructures are shifting away from familiar hardware-defined systems toward the new softwaredefined reality. Proprietary SAN-attached storage systems will slowly disappear from most environments, and be replaced by more efficient, modular hyperconverged building blocks.

REDUCING COMPLEXITY TO LOWER COSTS According to the Storage Networking Industry Association (SNIA), about 60 percent of Data Center CAPEX and OPEX is spent on SANs and SAN-attached storage.4 Hyperconvergence potentially eliminates all of these costs.

Intel Inside®. Powerful Solution Outside.

By converging storage, storage management, and the SAN at the hypervisor level and placing the SDS stack adjacent to the workloads, hyperconvergence reduces IT complexity, increases application performance, and delivers a dramatically more adaptive infrastructure. Storage-intensive applications become easier to deploy, manage, and scale than on traditional systems. This in turn reduces capital and operating costs — a critical consideration at a time when data is growing far faster than budgets. FASTER TO DEPLOY In a hyperconverged architecture, compute, storage, networking, and I/O functions are all collapsed into a single, software-defined infrastructure that can be deployed or scaled in a fraction of the time required for conventional hardware-defined, discrete elements. By eliminating purposebuilt or monolithic storage hardware and SAN switches, IT has far fewer physical elements to purchase and deploy and can do the job much more quickly.

Higher performance Applications access storage without traversing a SAN. Storage access is dramatically faster, response times are reduced by orders of magnitude, and applications run better when hyperconverged. High availability at lower cost Of course, Lenovo customers already enjoy higher reliability as evidenced by the TBR report.5 However, due to the greatly reduced CAPEX and OPEX costs of hyperconverged infrastructure, organizations and institutions can now implement high availability for those applications that previously could not be cost-justified for high availability. With far fewer discrete components in the architecture, hardware failures are less frequent, easier to manage, and more quickly resolved than with legacy systems.

Intel® Xeon® Processors.

For more information on SDS and hyperconvergence solutions, visit the Server & Storage FindIT page, LPN Government page, the LPN Education page or the Canadian LPN Velocity page.

PUBLIC SECTOR

Data efficiency One of the most important benefits of a hyperconverged data center is that the entire infrastructure is managed in a single system. This reduces workforce costs over the long term, as there is less need for specialized resources to manage disparate systems. Data protection Backup, recovery, and disaster recovery are easier to implement in software-defined and hyperconverged infrastructures — frequently eliminating the need to purchase and manage multiple disparate products.


LENOVO CONVERGED HX SERIES Lenovo Converged HX Series Powered by Nutanix deliver ready to deploy appliances designed to simplify every aspect of the data center infrastructure lifecycle — from procurement and deployment to management, scalability, and support. Customer benefits include: • Simplify IT infrastructure by integrating server, storage, and virtualization in a centrally managed appliance. • Reduce costs by implementing only what you need and then easily scale for future growth. AVAILABLE CONVERGED HX SERIES SOLUTIONS

POWERFUL PARTNERSHIPS FOR A HYPERCONVERGED DATA CENTER

• Lenovo Converged HX3500: Designed for compute-heavy workloads, this solution is ideal for VDI, web server, light databases, and smaller virtualization workloads.

Lenovo and Nutanix Lenovo’s global strategic partnership with Nutanix combines the industry-leading reliability of Lenovo hardware systems with software from the industry leader in hyperconverged appliances since 2011.

• Lenovo Converged HX5500: Designed for storage-heavy workloads, this solution can help customers with their workloads with large capacity needs, file servers, backup, Hadoop, and disaster recovery deployments (DR).

The partnership will reduce IT complexity and cost and increase greater efficiency and agility in data centers of all sizes through:

• Lenovo Converged HX7500: Designed for high performance, this solution can provide significant benefits for larger databases, MS SQL, Exchange, Oracle, and other I/O intensive workloads.

• A jointly developed and branded family of Lenovo appliances powered by Nutanix software

Nutanix Use Case: leading big-box retailer – test/dev and VDI A large electronics retailer with 2,000 locations in North America needed to support test/dev environments and thousands of VDI clients. Building these services on conventional SAN-attached storage architecture was far too expensive. Internal pilots using Amazon Web Services showed dramatic savings,6 but Amazon is the company’s chief competitor. A challenge was issued to IT managers to identify an on-premises solution that matched or beat AWS on both cost and performance criteria. After a detailed examination of several solutions, a private cloud infrastructure based on Nutanix was deployed. Nutanix demonstrated a higher performing and less costly alternative to AWS by: • Eliminating the expense of external SAN-attached or NAS storage and WAN connection costs • Improving performance by keeping the infrastructure local and deploying workloads and storage in the same servers • Eliminating AWS monthly subscription costs • Providing an easy-to-buy appliance model that allows the company to scale quickly and incrementally by merely adding additional nodes

• Sizable investments by both companies in joint platform development, performance engineering, and a dedicated and skilled salesforce Lenovo HX3500



PUBLIC SECTOR


modest compute and storage requirements

Lenovo and DataCore

DataCore Use Case: global auto parts distribution

System x3250 + StorMagic (VMware or Hyper-V)

Lenovo SDS and hyperconverged solutions powered by DataCore harness the full potential of today’s storage and server technology. With over 11,000 customers in production worldwide since 1999, including more than 1,600 in healthcare alone, DataCore is the global leader in storage virtualization and SDS.

A global auto parts distributor serving more than 10,000 retail locations runs a daily inventory replenishment application on traditional SAN-attached storage and AIX. Inventory turns and customer satisfaction are suffering as the “daily” replenishment job frequently takes more than 40 hours to complete.

• Large-scale distributed infrastructure

Any Lenovo System x, X6, or ThinkServer® with DataCore SANsymphony-V:

A two-server DataCore SDS solution using SanDisk Flash was deployed incrementally to solve the problem by:

• Delivers synchronous and asynchronous replication and CDP (continuous data protection) with unlimited distance between sites • Offers pooling and virtualization of heterogeneous SANattached storage as well as internal storage, JBOD, and DAS (including Lenovo S3200/S2200) • Can present physical storage “target” ports to existing SAN-attached application servers, including AIX (Power), Solaris (Oracle), and HP/UX (Itanium) • Is certified by Oracle, SAP, Microsoft, Citrix, and many others for high-end, enterprise workloads; SAP HANA Certified for TDI implementations • Is 100 percent hypervisor-agnostic; supported for VMware, Hyper-V, KVM, Citrix Xen, and Oracle Virtual Server • Has the demonstrated highest performance of any hyperconverged/SDS stack, world’s record SPC-1 — three times faster than EMC’s All-Flash VNX Array at about one-third the price7

• Application acceleration: Placing a two-server DataCore cluster with DRAM cache and Flash between the existing AIX servers and the existing EMC storage reduced the application run-time from 40 hours to four hours. • Hyperconverging and storage virtualization: The customer then ported the AIX application to Linux, and moved the database application into the DataCore SDS servers. By moving from SDS to hyperconverged, the application runtime was further reduced to less than one hour, enabling the replenishment job to run multiple times daily. Lenovo and StorMagic Lenovo and StorMagic’s Remote Office Branch Office (ROBO) solutions are available for entry-level, mid-range, and high-performance ROBO environments running VMware or Hyper-V. StorMagic has been the leader in hyperconverged for ROBO since 2008 and provides the only cost-effective, two-server solution recognized by VMware for ROBO environments. ThinkServer RS140 + StorMagic (Hyper-V only) • Two-server, high availability (HA) cluster that costs less than a single richly configured server

Intel Inside®. Powerful Solution Outside.

• Delivers enterprise-class remote management capabilities across hundreds or thousands of remote sites ThinkServer RD650 + StorMagic (Hyper-V only) • For storage-intensive HA applications • Meets tiered storage requirements in retail environments that combine video capture/analytics and mission-critical applications StorMagic Use Case: distributed retail/ROBO environment A large retail home-improvement chain with over 2,200 locations needed a highly available, small-footprint infrastructure at ultra-low cost. The company runs a number of applications onsite (point-of-sale, service scheduling, etc.) rather than rely on a central location. The company does not need high-performance, flash, or high-end storage capabilities, but it can’t risk operations ever being interrupted by a hardware failure or WAN outage. After implementing one of the world’s largest distributed, hyperconverged infrastructures, StorMagic software met the retailer’s goals by: • Providing a solution that requires only two servers for high availability (compared to three or four nodes for other solutions) • Eliminating an external iSCSI SAN array, reducing the amount of space, power, and maintenance needed for a highly available infrastructure • Delivering the lowest-cost hyperconverged solution

• Ideal for retail settings and remote environments with

Intel® Xeon® Processors.


PUBLIC SECTOR

Lenovo and SimpliVity Lenovo servers and SimpliVity’s OmniStack™ virtualization platform come together in a hyperconverged appliance to reduce data center complexity and costs. Lenovo System x3650 M5 + SimpliVity OmniStack (VMware only) • All IT services — compute, storage, and networking — in a single appliance • Real-time, inline deduplication, compression, and optimization implemented in SimpliVity’s FPGA hardware accelerator to conserve and maximize availability of server resources for customer application workloads • Built-in backup, disaster recovery, and WAN optimization SimpliVity Use Case: global automotive A global automotive company needed a data center refresh. The OmniStack with Lenovo System x3650 M5 hyperconverged solution offered something completely transformational and fit perfectly within the existing budget. The implementation of the SimpliVity solution on System x3650 M5 went flawlessly, and the company achieved superior ease of management for their two-man IT team.


LENOVO: IDEAL FOR HYPERCONVERGENCE

ARE YOUR CUSTOMERS READY FOR HYPERCONVERGED SOLUTIONS?

High performance Lenovo brings industry-leading performance to hyperconvergence in the data center. Our System x3650 M5 server platforms are rated No. 1 in x86 reliability 8 and achieve superior performance with Intel® Xeon® processors. The entire System x product line offers the lowest downtime of any x86 servers.9

• What are your overall infrastructure goals?

Built for the data center Lenovo offerings deliver enterprise-class features designed specifically to meet the demands of data centers. For example, industry-leading cooling in the System x3650 server supports a thermal range from 5° to 40°C. Best-of-breed interoperability There’s no “one size fits all” to optimize efficiency in the data center. That’s why Lenovo offers world-class data center systems built to seamlessly work with leading solutions providers. Customer satisfaction In survey after survey by Technology Business Research, Lenovo ranks No. 1 in customer satisfaction for System x.9 In the past five quarters, Lenovo’s six-month, one-time services and setup satisfaction score increased.10 Satisfaction continues post-purchase as Lenovo customers reported the highest satisfaction with customer care.11

• Are you considering a server, storage, or network refresh over the next 12 months? • What would it mean to your operations if your key applications performance were to increase five times? • Is data center consolidation part of a strategy to reduce the technology footprint? • Do you have a distributed enterprise environment? • What are the key workloads in the data center? • Do you need to simplify management to reduce operating costs? • Do you need to address reliability, scalability, or security? • Are you interested in lowering the cost of storage in your environment? • Do you need to improve your backup and recovery capabilities? Do you have strategies in place for disaster recovery and business continuity? • Do you have a Virtual Desktop Infrastructure (VDI)? If so, we can help lower its cost while improving performance. • Are you looking to improve your data protection plans? • Are you looking into an all-flash array to improve application performance? Are you exploiting Solid State Disks (SSDs)? Hybrid environments (SSDs + HDDs) have proven to very effective and affordable.



PUBLIC SECTOR


TO LEARN MORE Reach out to these internal contacts: Sales Specialists Chris Myhill [email protected] John Lytle [email protected] David Moore [email protected] Randy Mixon [email protected] Steve Riehl [email protected] Marco Rengan [email protected]

Technical SMEs Eric Wendel – Solutions architect, all offerings

[email protected]

Brandon Harrell – SimpliVity

[email protected]

Michael (Baker) Hull – Nutanix

[email protected]

For more information on Lenovo’s technology partners in hyperconvergence, visit: www.nutanix.com/lenovo

1.

Subramani, S. (14 September 2015). Hyper-Convergence - When Converged Systems Grow Up. IDC. Web.

2.

Adshead, A. (9 April 2014). Data set to grow 10-fold by 2020 as internet of things takes off. ComputerWeekly. Web.

3.

Carr, E. (21 April 2015). Riding the Waves of Big Data. Datanami. Web.

4.

Berman, S. (September 2013). Software-Defined Networks and the Future of Storage. [Presentation]. Web.

5.

Information Technology Intelligence Consulting. (July 2015). ITIC 2015-2016 Global Server Hardware, Server OS Reliability Report [Research Report]. Web.

6.

Barr, J. (13 November 2013). TCO Comparison: Amazon WorkSpaces and Traditional Virtual Desktop Infrastructure (VDI). Amazon Web Services, [Blog]. Web.

7.

DataCore Software Corporation. (30 November 2015). SPC Benchmark 1™ Full Disclosure Report. Storage Performance Council, SPC-1 V1.14. Web.

8.

Information Technology Intelligence Consulting. (May 2015). ITIC 2014-2015 Global Server Hardware, Server OS Reliability Report [Research Report]. Web.

9.

Technology Business Research, Inc. (5 August 2015). Corporate IT Buying Behavior & Customer Satisfaction Study: x86-based Servers, 1Q15 [Research Report]. Web.

10.

Technology Business Research, Inc. (25 January 2016). Corporate IT Buying Behavior & Customer Satisfaction Study: x86-based Servers, 4Q15 [Research Report].

11.

Technology Business Research, Inc. (25 January 2016). Corporate IT Buying Behavior & Customer Satisfaction Study: x86-based Servers, 4Q15 [Research Report].

www.stormagic.com/lenovo-stormagic-hyper-converged-sds-solution www.simplivity.com/products/omnistack-solution-with-lenovo-system-x3650-m5/ www.datacore.com/Partners/Current-Partners/TechnologyPartners/lenovo


© 2016 Lenovo. All rights reserved. Lenovo is not responsible for photographic or typographic errors. Lenovo, the Lenovo logo, ThinkServer, and System x are trademarks or registered trademarks of Lenovo. Intel, the Intel Logo, and Xeon are trademarks of Intel Corporation in the U.S. and/or other countries. All other trademarks are the property of their respective owners. Version 1.00, February 2016.
