Cracking the Silos of Custom Workflows

By Nicole Hemsoth

February 27, 2014

In high performance computing, the time-honored concept of creating tailored workflows to address complex requirements is nothing new. However, with the advent of new tools to analyze and process data—not to mention store, sort and manage it—traditional ways of thinking about HPC workflows are falling by the wayside in favor of new approaches that might help balance, stabilize and shatter siloed environments.

Moving beyond HPC specifically, there are certainly plenty of options for managing large-scale, diverse workflows that are designed specifically for cloud environments, and increasingly, for “big data” workflows that require orchestration between custom and commercial analytics stacks, involving hops from private or public clouds, into Hadoop and over to other analytics engines. The issue is, while there are dedicated tools for addressing workflow demands of HPC environments specifically (GridEngine, Platform, Adaptive, etc.) , or cloud environments in particular (OpenStack, etc.), some, including Adaptive Computing, argue that there are no tools that tackle HPC, cloud and the new range of big data opportunities all together—and in a way that’s primed for the custom workflow models that are so often found in some of the most complex enterprise and research datacenters.

In their experience with large organizations including NOAA, the Department of Defense and others, Adaptive Computing has had the opportunity to look under the hoods of some complicated engines for doing everything from oil exploration to addressing national security concerns. These users—and around 60% of those they recently surveyed across the public and private sector (beyond HPC exclusively) tended to have custom, homegrown workflows, which often leads to a host of problems, including a lack of flexibility to adopt new tools, time consumption spent on manually handling the complexity, and of course, overall inefficiency across the datacenter.

That 60% is a striking figure when one considers that the advent of new tools being considered to address the growing bevy of “big data” problems means more custom scripting and management of an already top-heavy stack. According to Adaptive Computing’s Jill King, this means the addition of more silos, which is exactly the opposite of what’s needed for mission-critical environments. Adaptive’s answer to this complexity is called Big Workflow, which for now means addressing these homegrown environments with a different type of glue than has been used to bind many of the HPC centers they’ve worked with over the last ten years with Moab.

King says that for many datacenter environments across the HPC and big data spectrum, the logjam happens at the important processing stage for complex data. This is currently very manual, time-consuming and laden with dependencies and, according to conversations they’ve had across multiple organizations, they’re finding a lot of “both over and under-utilized silos with long, complicated cues that simply aren’t efficient. “There’s a great need to unify, optimize and guarantee these environments,” King said.

Adaptive Computing senior architect, Daniel Hardman offered detail on Big Workflow, which is both an approach that requires custom tuning for homegrown environments via dedicated work with customer needs—as well as offering some new hooks for big data analytics tooling.

As you can see below, there are several separate silos, all governed at the top by what’s very often either a sophisticated homegrown or off-the-shelf system. Generally, says Hardman, there are not efficient ways of connecting the top level with the many silos below—and further, that top level framework can be connected across many parts of the datacenter and spectrum of needs; for example, that same level might be governing general business operations, a Hadoop cluster, an interface to a cloud pulling data off storage, and an HPC environment on top of all of that. It’s quite possible to do all of this—but it’s hardly efficient or manageable and leads to inflexibility given the processes that need to be worked in manually for custom workflows when new hooks are needed or something changes.

BigWorkflow1

“There a big need for an engine that’s capable of implementing a policy-based engine across these silos,” said Hardman. “We already sell Moab cloud suite, and a comparable product in HPC, and offer an integration with the Intel Hadoop side, and although we’ve not done a lot on public cloud it’s also possible. What’s needed then is something that makes it so a user’s custom glue can tap into some efficiency and automation–a new kind of coordination so that our Big Workflow coordinator can contact these silos and make things happen across those many silo boundaries.”

The goal of Big Workflow (the coordinator is not a product as much as an approach rooted in some new hooks they’ve provided to big data sources via the Intel Hadoop distro and more) is to provide all the logic in hard-coded scripts, which Hardman says can eliminate a lot of the duct tape with management across these silos. They’re still there, but the distinction between them is a lot less painful.

BigWorkflow2

“Most people in IT think about equilibrium—keep things humming-if things get broken, they get fixed. The problem is that big data is not friendly to that; it has an interesting relationship to storage in that it may not be convenient to think about those silo boundaries anymore. For example, I might have the same dataset, which begins its life in a public cloud, then I need to process and massage it in Hadoop, then perform some HPC computation on it after that, but that data may have all sorts of issues (privacy, regulatory, etc) I can’t just move it around or pretend that it’s local when it’s not. It has to be managed with policies that understand data movement, staging, management and more. Big data makes this Big Workflow coordination mandatory.”

For those familiar with Adaptive or using it already, there is the addition of the “data expert” concept, which is the smart part of the engine that has to be able to understand all about the data (lifecycle, forms through that lifecycle and its movement, size, who owns it, where can it be copied or not). This is coupled with some of the new automation for custom workflows found in the coordinator. As King explained, “we’re integrating with these custom workflows for now,  then we’ll branch out and make these standards but with so many people having custom workflows, we need to be able to provide something flexible.” In other words, Adaptive is using lessons learned with customers like Digital Globe (see a detailed writeup of how this works in action over at EnterpriseTech) to propel their work for custom environments and implement those lessons as standards to help broaden their APIs and reach into more areas, including tools like Tivoli, for instance.

This approach should resonate for folks in HPC and enterprise circles, according to IDC. “Our 2013 study revealed that a surprising two thirds of HPC sites are now performing big data analysis as part of their HPC workloads, as well as an uptick in combined uses of cloud computing and supercomputing,” said Chirag Dekate, Ph.D., research manager, High-Performance Systems at IDC. “As there is no shortage of big data to analyze and no sign of it slowing down, combined uses of cloud and HPC will occur with greater frequency, creating market opportunities for solutions such as Adaptive’s Big Workflow.”

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!

Researchers Scale COSMO Climate Code to 4888 GPUs on Piz Daint

October 17, 2017

Effective global climate simulation, sorely needed to anticipate and cope with global warming, has long been computationally challenging. Two of the major obstacles are the needed resolution and prolonged time to compute Read more…

By John Russell

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Cluster Competition coverage has come to its natural home: H Read more…

By Dan Olds

UCSD Web-based Tool Tracking CA Wildfires Generates 1.5M Views

October 16, 2017

Tracking the wildfires raging in northern CA is an unpleasant but necessary part of guiding efforts to fight the fires and safely evacuate affected residents. One such tool – Firemap – is a web-based tool developed b Read more…

By John Russell

HPE Extreme Performance Solutions

Transforming Genomic Analytics with HPC-Accelerated Insights

Advancements in the field of genomics are revolutionizing our understanding of human biology, rapidly accelerating the discovery and treatment of genetic diseases, and dramatically improving human health. Read more…

Exascale Imperative: New Movie from HPE Makes a Compelling Case

October 13, 2017

Why is pursuing exascale computing so important? In a new video – Hewlett Packard Enterprise: Eighteen Zeros – four HPE executives, a prominent national lab HPC researcher, and HPCwire managing editor Tiffany Trader Read more…

By John Russell

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Read more…

By Dan Olds

Intel Delivers 17-Qubit Quantum Chip to European Research Partner

October 10, 2017

On Tuesday, Intel delivered a 17-qubit superconducting test chip to research partner QuTech, the quantum research institute of Delft University of Technology (TU Delft) in the Netherlands. The announcement marks a major milestone in the 10-year, $50-million collaborative relationship with TU Delft and TNO, the Dutch Organization for Applied Research, to accelerate advancements in quantum computing. Read more…

By Tiffany Trader

Fujitsu Tapped to Build 37-Petaflops ABCI System for AIST

October 10, 2017

Fujitsu announced today it will build the long-planned AI Bridging Cloud Infrastructure (ABCI) which is set to become the fastest supercomputer system in Japan Read more…

By John Russell

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Intel Debuts Programmable Acceleration Card

October 5, 2017

With a view toward supporting complex, data-intensive applications, such as AI inference, video streaming analytics, database acceleration and genomics, Intel i Read more…

By Doug Black

OLCF’s 200 Petaflops Summit Machine Still Slated for 2018 Start-up

October 3, 2017

The Department of Energy’s planned 200 petaflops Summit computer, which is currently being installed at Oak Ridge Leadership Computing Facility, is on track t Read more…

By John Russell

US Exascale Program – Some Additional Clarity

September 28, 2017

The last time we left the Department of Energy’s exascale computing program in July, things were looking very positive. Both the U.S. House and Senate had pas Read more…

By Alex R. Larzelere

How ‘Knights Mill’ Gets Its Deep Learning Flops

June 22, 2017

Intel, the subject of much speculation regarding the delayed, rewritten or potentially canceled “Aurora” contract (the Argonne Lab part of the CORAL “ Read more…

By Tiffany Trader

Reinders: “AVX-512 May Be a Hidden Gem” in Intel Xeon Scalable Processors

June 29, 2017

Imagine if we could use vector processing on something other than just floating point problems.  Today, GPUs and CPUs work tirelessly to accelerate algorithms Read more…

By James Reinders

NERSC Scales Scientific Deep Learning to 15 Petaflops

August 28, 2017

A collaborative effort between Intel, NERSC and Stanford has delivered the first 15-petaflops deep learning software running on HPC platforms and is, according Read more…

By Rob Farber

Oracle Layoffs Reportedly Hit SPARC and Solaris Hard

September 7, 2017

Oracle’s latest layoffs have many wondering if this is the end of the line for the SPARC processor and Solaris OS development. As reported by multiple sources Read more…

By John Russell

US Coalesces Plans for First Exascale Supercomputer: Aurora in 2021

September 27, 2017

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, in Arlington, Va., yesterday (Sept. 26), it was revealed that the "Aurora" supercompute Read more…

By Tiffany Trader

Google Releases Deeplearn.js to Further Democratize Machine Learning

August 17, 2017

Spreading the use of machine learning tools is one of the goals of Google’s PAIR (People + AI Research) initiative, which was introduced in early July. Last w Read more…

By John Russell

GlobalFoundries Puts Wind in AMD’s Sails with 12nm FinFET

September 24, 2017

From its annual tech conference last week (Sept. 20), where GlobalFoundries welcomed more than 600 semiconductor professionals (reaching the Santa Clara venue Read more…

By Tiffany Trader

Graphcore Readies Launch of 16nm Colossus-IPU Chip

July 20, 2017

A second $30 million funding round for U.K. AI chip developer Graphcore sets up the company to go to market with its “intelligent processing unit” (IPU) in Read more…

By Tiffany Trader

Leading Solution Providers

Amazon Debuts New AMD-based GPU Instances for Graphics Acceleration

September 12, 2017

Last week Amazon Web Services (AWS) streaming service, AppStream 2.0, introduced a new GPU instance called Graphics Design intended to accelerate graphics. The Read more…

By John Russell

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

EU Funds 20 Million Euro ARM+FPGA Exascale Project

September 7, 2017

At the Barcelona Supercomputer Centre on Wednesday (Sept. 6), 16 partners gathered to launch the EuroEXA project, which invests €20 million over three-and-a-half years into exascale-focused research and development. Led by the Horizon 2020 program, EuroEXA picks up the banner of a triad of partner projects — ExaNeSt, EcoScale and ExaNoDe — building on their work... Read more…

By Tiffany Trader

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Cray Moves to Acquire the Seagate ClusterStor Line

July 28, 2017

This week Cray announced that it is picking up Seagate's ClusterStor HPC storage array business for an undisclosed sum. "In short we're effectively transitioning the bulk of the ClusterStor product line to Cray," said CEO Peter Ungaro. Read more…

By Tiffany Trader

Intel Launches Software Tools to Ease FPGA Programming

September 5, 2017

Field Programmable Gate Arrays (FPGAs) have a reputation for being difficult to program, requiring expertise in specialty languages, like Verilog or VHDL. Easin Read more…

By Tiffany Trader

IBM Advances Web-based Quantum Programming

September 5, 2017

IBM Research is pairing its Jupyter-based Data Science Experience notebook environment with its cloud-based quantum computer, IBM Q, in hopes of encouraging a new class of entrepreneurial user to solve intractable problems that even exceed the capabilities of the best AI systems. Read more…

By Alex Woodie

Intel, NERSC and University Partners Launch New Big Data Center

August 17, 2017

A collaboration between the Department of Energy’s National Energy Research Scientific Computing Center (NERSC), Intel and five Intel Parallel Computing Cente Read more…

By Linda Barney

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