Using In-Memory Data Grids for Global Data Integration

By Nicole Hemsoth

July 2, 2012

by Dr. William Bain, ScaleOut Software, Inc.

Introduction

By enabling extremely fast and scalable data access even under large and growing workloads, in-memory data grids (IMDGs) have proven their value in storing fast-changing application data. For example, Web server farms use IMDGs to hold and share large volumes of shopping carts under heavy Web loads. Applications in financial services use IMDGs to hold fast-changing stock trading data for processing orders or for quickly analyzing and responding to emerging market trends.

ScaleOut In Memory Servers

An increasing number of companies employ multiple data centers to distribute their workloads and mitigate the impact of catastrophic events such as earthquakes and floods. IMDGs can be used to complement disaster recovery strategies by continuously replicating changes to fast-changing grid-based data to remote sites. This enables fast recovery and resumption of processing without data loss after a disaster strikes.

The use of in-memory data grids has also created the opportunity for organizations to employ even more powerful global strategies for data sharing. As organizations work to efficiently access fast-changing data across multiple sites or scale their processing into the cloud, the need to quickly and seamlessly migrate data on demand has grown rapidly. For example, organizations that produce and store fast-changing data in multiple data centers need to be able to access and analyze data without regard to where it originates. Likewise, organizations that access the highly elastic resources of public clouds need an efficient way to restage data in the cloud for processing.

Because IMDGs are specifically designed to store fast-changing data, federating IMDGs across multiple sites and enabling seamless access to data among all federated sites provide an ideal solution to the challenge of global data access. The benefits are twofold. First, applications can efficiently access and update data simply by using the IMDG’s data access mechanisms without modification; the federated IMDGs handle all of the details of remote data access and coherent updating. Second, IMDGs provide the scalability and low latency required to enable applications to handle large workloads with fast responsiveness.

We describe the combined scenarios for data replication and sharing as global data integration. This article outlines how in-memory data grids easily can be deployed to implement key strategies for global data integration, and it describes the important benefits this technology brings to organizations with global reach.

Disaster Recovery

A solid disaster recovery strategy requires that if one data center goes offline, its workload can be handled by another healthy data center to avoid service interruptions. For this recovery strategy to be effective, changes to fast-changing application data must be continuously replicated to a remote site so that the site is immediately ready to handle the workload. An IMDG that includes site-to-site data replication to one or more IMDGs at remote sites can provide this important capability and thereby complement the data center’s other replication and recovery strategies. In addition, all data centers can be operated in a “live-live” configuration under normal operating conditions to make full use of all computing resources and avoid the need for an idle “stand-by” data center.

ScaleOut Disaster Recovery
 

Carefully integrating data replication technology into an IMDG’s software architecture enables it to deliver the performance and reliability needed to handle large, fast-changing workloads. It also enables this capability to be easily deployed and managed by IT administrators. ScaleOut GeoServer® DR from ScaleOut Software is an example of a technology that provides these capabilities.  Because it is designed to extend the scalable, highly available architecture of its underlying IMDG, ScaleOut StateServer® (SOSS), it automatically scales replication bandwidth as grid servers are added to handle growing workloads, and it automatically tolerates server failures without interrupting operations. Additionally, it provides management tools that allow IT staff to easily establish and monitor connections to remote sites.

Global Data Access

Beyond data replication for disaster recovery, global data integration provides a range of choices for federating data stored in IMDGs at multiple data centers and cloud sites. For example, multiple data centers can be integrated into a single virtual data grid to provide seamless access to data, regardless of where it is stored and where the access request originates. Also, multiple grids can be interconnected to provide automatic data migration and elastic scaling when needed.

ScaleOut Global Data Access To ensure that global data access can easily be integrated into applications, IMDGs can seamlessly incorporate global access into their data access mechanisms. This simplifies application design by making remote data access transparent and automatic. It also eliminates the need for applications to track where data is located and manually restage it for local access. As an example, ScaleOut GeoServer follows this approach by extending the APIs provided by ScaleOut StateServer to transparently access data on demand at a configured set of remote sites; all grid accesses proceed as if data were located in the application’s local IMDG. ScaleOut GeoServer automatically searches remote IMDGs for missing data and copies it into the local IMDG as needed.

“Mostly Read” Access

ScaleOut GeoServer gives applications fine-grained control over data sharing to ensure efficient use of wide area networks (WANs) and to support various usage models. In one important use case, described as “mostly read” access, applications primarily need to access certain remote data but not perform updates on that data. This type of remotely accessed data is typically static or slowly changing so that local copies only need infrequent refresh over the WAN. Examples could include product pricing information for Web sites or portfolio holdings in financial services.

ScaleOut GeoServer implements mostly-read access by creating a local copy of remotely accessed data and allowing the application to specify a policy for refreshing it. The use of a local copy keeps local reads fast and minimizes WAN usage. Individual data objects can be marked by the application either to be updated periodically or to be updated when a change occurs at the remote site. Called coherency policies, these rules allow applications to tailor WAN usage to the characteristics of the data being remotely accessed.

An example of mostly read access, consider a wealth management application that needs to update its portfolios with periodic price changes; prices for different investments are held in multiple data grids around the world. The application can use global data access to obtain and efficiently track prices, with updates flowing into its local IMDG at the frequency required by the application. Also, to minimize WAN usage, only the prices of investments specifically needed by the application are retrieved over the WAN.

ScaleOut Mostly Read Access 

“Read/Write” Access

In a second important use case called “read/write” access, remotely accessed data needs to be accessed and then updated, and updates by different sites need to be carefully synchronized.  Examples include shopping carts in a Web site or financial portfolios being managed (not just examined) at remote sites. These data types can be fast-changing, and it is imperative to synchronize updates to avoid corrupting vital application data.

To synchronize updates, data must migrate from site to site on demand and avoid the use of local copies which could become out of date. ScaleOut GeoServer implements data migration and read/write access by transparently incorporating it into the IMDG’s existing distributed locking mechanism, which has been extended to span multiple sites. The IMDG automatically migrates ownership of data from a remote site when it is locked for reading by the application. This ensures that updates are always performed locally and at exactly one site at a time. The application does not have to manually restage data across sites nor provide its own mechanism for global data synchronization.

As an example, consider a premise-hosted ecommerce Web farm that needs to scale into the cloud to handle high seasonal demand. To accomplish this, the Web site’s administrator reconfigures the IP load-balancer to distribute Web requests across both on-premise and cloud-based Web servers; this procedure is sometimes called “cloud bursting.” By using an IMDG capable of global data integration, all Web servers transparently and coherently retrieve and update shopping carts within a single, virtualized IMDG spanning both sites. The following diagram illustrates this scenario using ScaleOut StateServer (“SOSS”) IMDGs at both sites and ScaleOut GeoServer to provide automatic data migration.

ScaleOut Read/Write Access 

Combining Data Replication and Global Data Access

It is often useful to combine the capabilities described above for global data integration to simultaneously address multiple requirements. For example, two central data centers which hold data accessed by satellite data centers can use data replication for disaster recovery purposes. Both could handle live traffic as described above, but in the case of a data center failure all traffic is routed to the healthy data center. Applications running in satellite data centers can use global data access to retrieve and/or update data held in the two central data centers. These applications can access data from either data center and transparently receive it even if one of the central data centers goes down. As illustrated in the following diagram, this configuration demonstrates the power and flexibility of global data integration.

ScaleOut Benefits of a Virtual Data Grid 

Benefits of a Virtual Data Grid

As we have seen, the goals of global data integration are to replicate data for disaster recovery and to enable applications to transparently access data across multiple sites as needed. ScaleOut GeoServer’s implementation of global data integration accomplishes these goals by creating a virtual data grid that seamlessly federates in-memory data grids across multiple sites. This enables application developers to write programs which access all shared data from a single (local) IMDG, leaving the IMDG to implement the details of remote access and synchronization. After a minimal amount of configuration to connect to remote sites, changes to add or remove grid servers in any data center do not affect configuration of the virtual data grid. The virtual data grid is able to withstand and recover from WAN interruptions and other failure conditions without affecting applications.

This article has illustrated the power of global data integration to extend the reach of applications that manage data spanning multiple data centers. As we have seen, in-memory data grids (IMDGs) provide a fast, scalable storage repository for application data. Their mechanisms can be transparently extended to enable data replication for disaster recovery and global access to data held at remote sites. These capabilities open up important new scenarios for globally distributed applications and simplify their implementation. Now applications can seamlessly access data worldwide and extend their processing into the cloud to handle peak workloads. Managing geographically distributed data has never been easier.

 

Dr. William L. Bain is founder and CEO of ScaleOut Software, Inc. Bill has a Ph.D. in electrical engineering/parallel computing from Rice University, and he has worked at Bell Labs research, Intel, and Microsoft. Bill founded and ran three start-up companies prior to joining Microsoft. In the most recent company (Valence Research), he developed a distributed Web load-balancing software solution that was acquired by Microsoft and is now called Network Load Balanc­ing within the Windows Server operating system. Dr. Bain holds several patents in computer architecture and distributed computing. As a member of the Seattle-based Alliance of Angels, Dr. Bain is actively involved in entrepreneurship and the angel community.

www.scaleoutsoftware.com

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

ExxonMobil, NCSA, Cray Scale Reservoir Simulation to 700,000+ Processors

February 17, 2017

In a scaling breakthrough for oil and gas discovery, ExxonMobil geoscientists report they have harnessed the power of 717,000 processors – the equivalent of 22,000 32-processor computers – to run complex oil and gas reservoir simulation models. Read more…

By Doug Black

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

HPE Extreme Performance Solutions

Object Storage is the Ideal Storage Method for CME Companies

The communications, media, and entertainment (CME) sector is experiencing a massive paradigm shift driven by rising data volumes and the demand for high-performance data analytics. Read more…

Weekly Twitter Roundup (Feb. 16, 2017)

February 16, 2017

Here at HPCwire, we aim to keep the HPC community apprised of the most relevant and interesting news items that get tweeted throughout the week. Read more…

By Thomas Ayres

Alexander Named Dep. Dir. of Brookhaven Computational Initiative

February 15, 2017

Francis Alexander, a physicist with extensive management and leadership experience in computational science research, has been named Deputy Director of the Computational Science Initiative at the U.S. Read more…

Here’s What a Neural Net Looks Like On the Inside

February 15, 2017

Ever wonder what the inside of a machine learning model looks like? Today Graphcore released fascinating images that show how the computational graph concept maps to a new graph processor and graph programming framework it’s creating. Read more…

By Alex Woodie

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

Cray Posts Best-Ever Quarter, Visibility Still Limited

February 10, 2017

On its Wednesday earnings call, Cray announced the largest revenue quarter in the company’s history and the second-highest revenue year. Read more…

By Tiffany Trader

HPC Cloud Startup Launches ‘App Store’ for HPC Workflows

February 9, 2017

“Civilization advances by extending the number of important operations which we can perform without thinking about them,” Read more…

By Tiffany Trader

Intel and Trump Announce $7B for Fab 42 Targeting 7nm

February 8, 2017

In what may be an attempt by President Trump to reset his turbulent relationship with the high tech industry, he and Intel CEO Brian Krzanich today announced plans to invest more than $7 billion to complete Fab 42. Read more…

By John Russell

For IBM/OpenPOWER: Success in 2017 = (Volume) Sales

January 11, 2017

To a large degree IBM and the OpenPOWER Foundation have done what they said they would – assembling a substantial and growing ecosystem and bringing Power-based products to market, all in about three years. Read more…

By John Russell

US, China Vie for Supercomputing Supremacy

November 14, 2016

The 48th edition of the TOP500 list is fresh off the presses and while there is no new number one system, as previously teased by China, there are a number of notable entrants from the US and around the world and significant trends to report on. Read more…

By Tiffany Trader

Lighting up Aurora: Behind the Scenes at the Creation of the DOE’s Upcoming 200 Petaflops Supercomputer

December 1, 2016

In April 2015, U.S. Department of Energy Undersecretary Franklin Orr announced that Intel would be the prime contractor for Aurora: Read more…

By Jan Rowell

D-Wave SC16 Update: What’s Bo Ewald Saying These Days

November 18, 2016

Tucked in a back section of the SC16 exhibit hall, quantum computing pioneer D-Wave has been talking up its new 2000-qubit processor announced in September. Forget for a moment the criticism sometimes aimed at D-Wave. This small Canadian company has sold several machines including, for example, ones to Lockheed and NASA, and has worked with Google on mapping machine learning problems to quantum computing. In July Los Alamos National Laboratory took possession of a 1000-quibit D-Wave 2X system that LANL ordered a year ago around the time of SC15. Read more…

By John Russell

Enlisting Deep Learning in the War on Cancer

December 7, 2016

Sometime in Q2 2017 the first ‘results’ of the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) will become publicly available according to Rick Stevens. He leads one of three JDACS4C pilot projects pressing deep learning (DL) into service in the War on Cancer. Read more…

By John Russell

HPC Startup Advances Auto-Parallelization’s Promise

January 23, 2017

The shift from single core to multicore hardware has made finding parallelism in codes more important than ever, but that hasn’t made the task of parallel programming any easier. Read more…

By Tiffany Trader

IBM Wants to be “Red Hat” of Deep Learning

January 26, 2017

IBM today announced the addition of TensorFlow and Chainer deep learning frameworks to its PowerAI suite of deep learning tools, which already includes popular offerings such as Caffe, Theano, and Torch. Read more…

By John Russell

CPU Benchmarking: Haswell Versus POWER8

June 2, 2015

With OpenPOWER activity ramping up and IBM’s prominent role in the upcoming DOE machines Summit and Sierra, it’s a good time to look at how the IBM POWER CPU stacks up against the x86 Xeon Haswell CPU from Intel. Read more…

By Tiffany Trader

Leading Solution Providers

Nvidia Sees Bright Future for AI Supercomputing

November 23, 2016

Graphics chipmaker Nvidia made a strong showing at SC16 in Salt Lake City last week. Read more…

By Tiffany Trader

BioTeam’s Berman Charts 2017 HPC Trends in Life Sciences

January 4, 2017

Twenty years ago high performance computing was nearly absent from life sciences. Today it’s used throughout life sciences and biomedical research. Genomics and the data deluge from modern lab instruments are the main drivers, but so is the longer-term desire to perform predictive simulation in support of Precision Medicine (PM). There’s even a specialized life sciences supercomputer, ‘Anton’ from D.E. Shaw Research, and the Pittsburgh Supercomputing Center is standing up its second Anton 2 and actively soliciting project proposals. There’s a lot going on. Read more…

By John Russell

Container App ‘Singularity’ Eases Scientific Computing

October 20, 2016

HPC container platform Singularity is just six months out from its 1.0 release but already is making inroads across the HPC research landscape. It's in use at Lawrence Berkeley National Laboratory (LBNL), where Singularity founder Gregory Kurtzer has worked in the High Performance Computing Services (HPCS) group for 16 years. Read more…

By Tiffany Trader

Dell Knights Landing Machine Sets New STAC Records

November 2, 2016

The Securities Technology Analysis Center, commonly known as STAC, has released a new report characterizing the performance of the Knight Landing-based Dell PowerEdge C6320p server on the STAC-A2 benchmarking suite, widely used by the financial services industry to test and evaluate computing platforms. The Dell machine has set new records for both the baseline Greeks benchmark and the large Greeks benchmark. Read more…

By Tiffany Trader

IDG to Be Bought by Chinese Investors; IDC to Spin Out HPC Group

January 19, 2017

US-based publishing and investment firm International Data Group, Inc. (IDG) will be acquired by a pair of Chinese investors, China Oceanwide Holdings Group Co., Ltd. Read more…

By Tiffany Trader

What Knights Landing Is Not

June 18, 2016

As we get ready to launch the newest member of the Intel Xeon Phi family, code named Knights Landing, it is natural that there be some questions and potentially some confusion. Read more…

By James Reinders, Intel

KNUPATH Hermosa-based Commercial Boards Expected in Q1 2017

December 15, 2016

Last June tech start-up KnuEdge emerged from stealth mode to begin spreading the word about its new processor and fabric technology that’s been roughly a decade in the making. Read more…

By John Russell

Intel and Trump Announce $7B for Fab 42 Targeting 7nm

February 8, 2017

In what may be an attempt by President Trump to reset his turbulent relationship with the high tech industry, he and Intel CEO Brian Krzanich today announced plans to invest more than $7 billion to complete Fab 42. Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This