Many people associate high-performance computing (HPC), also known as supercomputing, with far-reaching government-funded research or consortia-led efforts to map the human genome or to pursue the latest cancer cure.

But HPC can also be tapped to advance more traditional business outcomes — from fraud detection and intelligent operations to helping advance digital transformation. The challenge: making complex compute-intensive technology accessible for mainstream use.

As companies digitally transform and steer toward becoming data-driven businesses, there is a need for increased computing horsepower to manage and extract business intelligence and drive data-intensive workloads at scale. The rise of artificial intelligence (AI), machine learning (ML), and real-time analytics applications, often deployed at the edge, can utilize HPC resources to unlock insights from data and efficiently run increasingly large and more complex models and simulations.

The convergence of HPC with AI-based analytics is impacting nearly every industry and across a wide range of applications, including space exploration, drug discovery, financial modeling, automotive design, and systems engineering.

“HPC is becoming a utility in our lives — people aren’t thinking about what it takes to design this tire, validate a chip design, parse and analyze customer preferences, do risk management, or build a 3D structure of the COVID-19 virus,” notes Max Alt, distinguished technologist and director of Hybrid HPC at HPE. “HPC is everywhere, but you don’t think about it, because it’s hidden at the core.”

HPC’s scalable architecture is particularly well suited for AI applications, given the nature of computation required and the unpredictable growth of data associated with these workflows. HPC’s use of graphics-processing-unit (GPU) parallel processing power — coupled with its simultaneous processing of compute, storage, interconnects, and software — raises the bar on AI efficiencies. At the same time, such applications and workflows can operate and scale more readily.

Even with widespread usage, there is more opportunity to leverage HPC for better and faster outcomes and insights. HPC architecture — typically clusters of CPU and GPUs working in parallel and connected to a high-speed network and data storage system — is expensive, requiring a significant capital investment. HPC workloads are typically associated with vast data sets, which means that public cloud might be an expensive option due to requirements regarding latency and performance issues. In addition, data security and data gravity concerns often rule out public cloud.

Another major barrier to more widespread deployment: a lack of in-house specialized expertise and talent. HPC infrastructure is far more complex than traditional IT infrastructure, requiring specialized skills for managing, scheduling, and monitoring workloads. “You have tightly coupled computing with HPC, so all of the servers need to be well synchronized and performing operations in parallel together,” Alt explains. “With HPC, everything needs to be in sync, and if one node goes down, it can fail a large, expensive job. So you need to make sure there is support for fault tolerance.”

HPE GreenLake for HPC Is a Game Changer

An as-a-service approach can address many of these challenges and unlock the power of HPC for digital transformation. HPE GreenLake for HPC enables companies to unleash the power of HPC without having to make big up-front investments on their own. This as-a-service-based delivery model enables enterprises to pay for HPC resources based on the capacity they use. At the same time, it provides access to third-party experts who can manage and maintain the environment in a company-owned data center or colocation facility while freeing up internal IT departments.

“The trend of consuming what used to be a boutique computing environment now as-a-service is growing exponentially,” Alt says.

HPE GreenLake for HPC bundles the core components of an HPC solution (high-speed storage, parallel file systems, low-latency interconnect, and high-bandwidth networking) in an integrated software stack that can be assembled to meet an organization’s specific workload needs.

As part of the HPE GreenLake edge-to-cloud platform, HPE GreenLake for HPC gives organizations access to turnkey and easily scalable HPC capabilities through a cloud service consumption model that’s available on-premises. The HPE GreenLake platform experience provides transparency for HPC usage and costs and delivers self-service capabilities; users pay only for the HPC resources they consume, and built-in buffer capacity allows for scalability, including unexpected spikes in demand. HPE experts also manage the HPC environment, freeing up IT resources and delivering access to the specialized performance tuning, capacity planning, and life cycle management skills.

To meet the needs of the most demanding compute and data-intensive workloads, including AI and ML initiatives, HPE has turbocharged HPE GreenLake for HPC with purpose-built HPC capabilities. Among the more notable features are expanded GPU capabilities, including NVIDIA Tensor Core models; support for high-performance HPE Parallel File System Storage; multicloud connector APIs; and HPE Slingshot, a high-performance Ethernet fabric designed to meet the needs of data-intensive AI workloads. HPE also released lower entry points to HPC to make the capabilities more accessible for customers looking to test and scale workloads.

As organizations pursue HPC capabilities, they should consider the following:

Stop thinking of HPC in terms of a specialized boutique technology; think of it more as a common utility used to drive business outcomes.Look for HPC options that are supported by a rich ecosystem of complementary tools and services to drive better results and deliver customer excellence.Evaluate the HPE GreenLake for HPC model. Organizations can dial capabilities up and down, depending on need, while simplifying access and lowering costs.

HPC horsepower is critical, as data-intensive workloads, including AI, take center stage. An as-a-service model democratizes what’s traditionally been out of reach for most, delivering an accessible path to HPC while accelerating data-first business.

For more information, visit https://www.hpe.com/us/en/greenlake/high-performance-compute.html

High-Performance Computing

IT leaders seeking to derive business value from the data their companies collect face myriad challenges. Perhaps the least understood is the lost opportunity of not making good on data that is created, and often stored, but seldom otherwise interacted with.

This so-called “dark data,” named after the dark matter of physics, is information routinely collected in the course of doing business: It’s generated by employees, customers, and business processes. It’s generated as log files by machines, applications, and security systems. It’s documents that must be saved for compliance purposes, and sensitive data that should never be saved, but still is.

According to Gartner, the majority of your enterprise information universe is composed of “dark data,” and many companies don’t even know how much of this data they have. Storing it increases compliance and cybersecurity risks, and, of course, doing so also increases costs.

Figuring out what dark data you have, where it is kept, and what information is in it is an essential step to ensuring the valuable parts of this dark data are secure, and those that shouldn’t be kept are deleted. But the real advantage to unearthing these hidden pockets of data may be in putting it to use to actually benefit the business.

But mining dark data is no easy task. It comes in a wide variety of formats, can be completely unformatted, locked away in scanned documents or audio or video files, for example.

Here is a look at how some organizations are transforming dark data into business opportunities, and what advice industry insiders have for IT leaders looking to leverage dark data.

Coded audio from race car drivers

For five years, Envision Racing has been collecting audio recordings from more than 100 Formula E races, each with more than 20 drivers.

“The radio streams are available on open frequencies for anyone to listen to,” says Amaresh Tripathy, global leader of analytics at Genpact, a consulting company that helped Envision Racing make use of this data.

Previously the UK-based racing team’s race engineers tried to use these audio transmissions in real-time during races, but the code names and acronyms drivers used made it difficult to figure out what was being said and how it could be made use of, as understanding what other drivers were saying could help Envision Racing’s drivers with their racing strategy, Tripathy says.

“Such as when to use the attack mode. When to overtake a driver. When to apply brakes,” he says.

Envision Racing was also collecting sensor data from its own cars, such as from tires, batteries, and breaks, and purchasing external data from vendors, such as wind speed and precipitation.

Genpact and Envision Racing worked together to unlock the value of these data streams, making use of natural language processing to build deep learning models to analyze them. The process took six months, from preparing the data pipeline, to ingesting the data, to filtering out noise, to deriving meaningful conversations.

Tripathy says humans take five to ten seconds to figure out what they’re listening to, a delay that made the radio communications irrelevant. Now, thanks to the AI model’s predictions and insights, they can now respond in one to two seconds.

In July, at the ABB FIA Formula E World Championship in New York, the Envision Racing team took first and third places, a result Tripathy credits to making use of what was previously dark data.

Dark data gold: Human-generated data

Envision Racing’s audio files are an example of dark data generated by humans, intended for consumption by other humans — not by machines. This kind of dark data can be extremely useful for enterprises, says Kon Leong, co-founder and CEO of ZL Technologies, a data archiving platform provider.

“It is incredibly powerful for understanding every element of the human side of the enterprise, including culture, performance, influence, expertise, and engagement,” he says. “Employees share absolutely massive amounts of digital information and knowledge every single day, yet to this point it’s been largely untapped.”

The information contained in emails, messages, and files can help organizations derive insights such as who are the most influential people are in the organization. “Eighty percent of company time is spent communicating. Yet analytics often deals with data that only reflects 1% of our time spent,” Leong says.

Processing human-generated unstructured data is uniquely challenging. Data warehouses aren’t typically set up to handle these communications, for example. Moreover, collecting these communications can create new issues for companies to deal with, having to do with compliance, privacy, and legal discovery.

“These governance capabilities are not present in today’s concept of a data lake, and in fact by collecting data into a data lake, you create another silo which increases privacy and compliance risks,” Leong says.

Instead companies can also leave this data where it currently resides, simply adding a layer of indexing and metadata for searchability. Leaving the data in place will also keep it within existing compliance structures, he says.

Effective governance is key

Another approach to handling dark data of questionable value and origin is to start with traceability.

“It’s a positive development in the industry that dark data is now recognized as an untapped resource that can be leveraged,” says Andy Petrella, author of Fundamentals of Data Observability, currently available in pre-release form from O’Reilly. Petrella is also the founder of data observability provider Kensu.

“The challenge with utilizing dark data is the low levels of confidence in it,” he says, in particular around where and how the data is collected. “Observability can make data lineage transparent, hence traceable. Traceability enables data quality checks that lead to confidence in employing these data to either train AI models or act on the intelligence that it brings.”

Chuck Soha, managing director at StoneTurn, a global advisory firm specializing in regulatory, risk, and compliance issues, agrees that the common approach to tackling dark data — throwing everything into a data lake — poses significant risks.

This is particularly true in the financial services industry, he says, where companies have been sending data into data lakes for years. “In a typical enterprise, the IT department dumps all available data at their disposal into one place with some basic metadata and creates processes to share with business teams,” he says.

That works for business teams that have the requisite analytics talent in-house or that bring in external consultants for specific use cases. But for the most part these initiatives are only partially successful, Soha says.

“CIOs transformed from not knowing what they don’t know to knowing what they don’t know,” he says.

Instead, companies should begin with data governance to understand what data there is and what issues it might have, data quality chief among them.

“Stakeholders can decide whether to clean it up and standardize it, or just start over with better information management practices,” Soha says, adding that investing in extracting insights from data that contains inconsistent or conflicting information would be a mistake.

Soha also advises connecting the dots between good operational data already available inside individual business units. Figuring out these relationships can create rapid and useful insights that might not require looking at any dark data right away, he says. “And it might also identify gaps that could prioritize where in the dark data to start to look to fill those gaps in.”

Finally, he says, AI can be very useful in helping make sense of the unstructured data that remains. “By using machine learning and AI techniques, humans can look at as little as 1% of dark data and classify its relevancy,” he says. “Then a reinforcement learning model can quickly produce relevancy scores for the remaining data to prioritize which data to look at more closely.”

Using AI to extract value

Common AI-powered solutions for processing dark data include Amazon’s Textract, Microsoft’s Azure Cognitive Services, and IBM’s Datacap, as well as Google’s Cloud Vision, Document, AutoML, and NLP APIs.

In Genpact’s partnership with Envision Racing, Genpact coded the machine learning algorithms in-house, Tripathy says. This required knowledge of Docker, Kubernetes, Java, and Python, as well as NLP, deep learning, and machine learning algorithm development, he says, adding that an MLOps architect managed the complete process.

Unfortunately, these skills are hard to come by. In a report released last fall by Splunk, only 10% to 15% of more than 1,300 IT and business decision makers surveyed said their organizations are using AI to solve the dark data problem. Lack of necessary skills was a chief obstacle to making use of dark data, second only to the volume of the data itself.

A problem (and opportunity) on the rise

In the meantime, dark data remains a mounting trove of risk — and opportunity. Estimates of the portion of enterprise data that is dark vary from 40% to 90%, depending on industry.

According to a July report from Enterprise Strategy Group, and sponsored by Quest, 47% of all data is dark data, on average, with a fifth of respondents saying more than 70% of their data is dark data. Splunk’s survey showed similar findings, with 55% of all enterprise data, on average, being dark data, and a third of respondents saying that 75% or more of their organization’s data is dark.

And the situation is likely to get worse before it gets better, as 60% of respondents say that more than half of the data in their organization is not captured at all and much of it is not even understood to exist. As that data is found and stored, the amount of dark data is going to continue to go up.

It’s high time CIOs put together a plan on how to deal with it — with an eye toward making the most of any dark data that shows promise in creating new value for the business.

Analytics, Data Management, Data Science

A modern, agile IT infrastructure has become the critical enabler for success, allowing organizations to unlock the potential of new technologies such as AI, analytics, and automation. Yet modernization journeys are often bumpy; IT leaders must overcome barriers such as resistance to change, management complexity, high costs, and talent shortages.

Those successful in their modernization endeavors can expect significant business gains. In Ampol’s case, the transport fuels provider enjoyed enhanced operational efficiency, business agility, and maximized service uptimes.

A vision for transformation, hampered by legacy

Ampol had a clear goal: intelligent operations for improved service reliability, increased agility, and reduced cost. To achieve this, Ampol created a vision centered on “uplifting and modernizing existing cloud environment and practices,” according to Lindsay Hoare, Ampol’s Head of Technology.

This meant having enterprise-wide visibility and environment transparency for real-time updates, modernizing its environment management capabilities with cloud-based and cloud-ready tools, building the right capabilities and skillsets for the cloud, redesigning the current infrastructure into a cloud-first one, and leveraging automation for enhanced operations.  

While Ampol had most workloads in the cloud, it is still highly dependent on its data center. This meant added complexity to infrastructure networking and management, which in turn drove up maintenance and management costs. The need for human intervention across the environment further increased the risk of error and resultant downtime. Its ambition to enable automation across the entire enterprise, at that point in time, felt unattainable as it lacked the technical expertise and capabilities to do so.

Realizing its ambitions with the right partner

Ampol knew it was not able to modernize its enterprise and bridge the ambition gap alone. It then turned to Accenture. “We needed a partner with a cloud mindset, one that could cover the technological breadth at which Ampol operates,” said Hoare. “Hence why we turned to Accenture, with whom we’ve built a strong partnership that has spanned over a decade.”

Accenture has been helping Ampol in its digital transformation journey across many aspects of its IT operations and as such has a deep understanding of Ampol’s automation ambitions.

“We brought to the table our AIOps capability that leverages automation, analytics, and AI for intelligent operations. Through our ongoing work with Ampol, we were able to accelerate cloud adoption alongside automation implementation, reducing implementation and deployment time,” said Duncan Eadie, Accenture’s Managing Director of Cloud, Infra, and Engineering for AAPAC.

Reaping business benefits through intelligent operations

Through its collaboration with Accenture, Ampol was able to realize its vision for intelligent operations which then translates to business benefits.

Visualization and monitoring

Ampol can now quickly pinpoint incidents to reduce the time to resolve. Recently, a device failure impacted Ampol’s retail network and service stations, but a map-based visualization of the network allowed engineers to identify the device and switch over to the secondary within the hour: an 85% improvement in downtime reduction.

Self-healing capabilities

Intelligent operations not only detect failures but also attempt to resolve them independently and create incidents for human intervention only when basic resolution is unsuccessful. As a result, Ampol’s network incidents have been reduced by 40% while business-impacting retail incidents are down by half.

Automating mundane tasks

Automation now regularly takes care of mundane and routine tasks such as patching, updates, virtual machine builds, and software installs. This frees up employees’ time that is otherwise spent on maintenance, enabling them to innovate and add real business value through working on more strategic assignments and business growth.

Future-proofing

As Ampol focuses on the global energy transition, it is investing in new energy solutions in a highly dynamic environment. A cloud-first infrastructure removes complexity, increases the levels of abstraction, and offers greater leverage of platform services, enabling agility and responsiveness. The right architecture and security zoning facilitate critical business-led experimentation and innovation to ensure Ampol continues to place at the front of the pack.

As IT infrastructure becomes a critical enabler across industries, organizations are compelled to embrace modernization. While significant roadblocks exist, a clear vision and the right partner can help overcome challenges and unlock the potential of the cloud, AI and analytics, and automation, to be a true game-changer.

“This is a long journey,” says Hoare, “we’ve been at it for years now… It needs drive and tenacity. But when you get there, you’ll be in a great place.”

Learn more about getting started with a modern infrastructure here.

Cloud Management, Digital Transformation