On the Spack Track @SC19

By Elizabeth Leake

December 5, 2019

At the annual supercomputing conference, SC19 in Denver, Colorado, there were Spack events each day of the conference. As a reflection of its grassroots heritage, nine sessions were planned by more than a dozen thought leaders from seven organizations, including three U.S. national Department of Energy (DOE) laboratories and Sylabs, the company behind Singularity. Thirteen thousand six hundred conference attendees had the chance to learn about Spack from two meet-and-greets, three birds-of-a-feather (BoF) meetings, three papers, and more.

What is Spack?

Spack is an open-source scientific software package manager for high-performance computing (HPC) environments, MacOS and Linux platforms. It simplifies an otherwise tedious and time-consuming task. There are currently more than 3,600 packages in the Spack library; with 2,300 active users and over 480 contributors from national labs, academia and industry, the number of packages is constantly growing and being improved with community input.

From the Spack website, spack.io, “Packages can be built with multiple versions, configurations, platforms, and compilers, and all of these builds can coexist on the same machine. It isn’t tied to a particular language; a software stack can be built in Python or R, link to libraries written in C, C++, or Fortran, easily swap compilers, and target specific microarchitectures. Spack can be used to install without root in a home directory, to manage shared installations and modules, or to build combinatorial versions of software for testing.” Packages are templated so users can easily tune for the host environment.

Ninety attended the November 21 BoF led by Todd Gamblin (Senior Principal Member of Technical Staff at Lawrence Livermore National Laboratory (LLNL), middle); Adam Stewart (UIUC); Massimiliano Culpo (Sylabs, Inc.); Greg Becker (LLNL, right); and Peter Scheibel (LLNL, left).

Gamblin leads the Spack development team at LLNL with fellow computer scientists Becker, Scheibel, Tamara Dahlgren, Gregory Lee, and Matt Legendre. Co-developers are from Argonne National Laboratory; Columbia University; École polytechnique fédérale de Lausanne; Fermi National Accelerator Laboratory, Iowa State University; Kitware, Inc.; NASA Goddard Institute for Space Studies, the Center for Climate Systems Research; the National Energy Research Scientific Computing Center; Perimeter Institute; the University of Hamburg, the University of Illinois at Urbana-Champaign; and the University of Iowa.

With widespread adoption among DOE labs, it’s not surprising that the high-energy physics community based at Fermilab in the U.S. and at CERN (European Organization for Nuclear Research) in Switzerland is replacing a longstanding solution with end-to-end tooling built around Spack.

The BoF incorporated a real-time audience polling tool called Glisser. Participant questions and answers populated pie charts and word clouds, which helped to identify development priorities for the future.

When asked what people loved about Spack, most cited convenience in dealing with dependencies. DOE shared the following victories:

  • At LLNL, Spack was used to automate the radiation hydrodynamics code, ARES, with 40 dependency libraries on a dozen compilers running on Blue Gene/Q and commodity Linux clusters. Porting time on a new platform was reduced from two weeks to three hours.
  • At Oak Ridge National Laboratory, Spack reduced deployment time on Summit, the world’s fastest publicly-ranked supercomputer, from two weeks to twelve hours. Summit’s stack of 1,300 packages can be built overnight.
  • DOE’s Exascale Computing Project (ECP) is tasked with ensuring exascale-readiness for the U.S. lab complex. Its software stack is large and complex; a brand-new class of 90 applications is managed with Spack so that users will be able to easily incorporate them in many different environments. Security features were added to the open-source GitLab product that integrate with each center’s identity management and schedulers, such as SLURM and LSF.
  • The Spack development team is in the process of democratizing a pipeline of software management across six DOE labs; automated package builds with both public and private repositories will work with dozens of systems.

Spack is used with Germany’s SuperMUC-NG at the Leibniz Supercomputing Centre of the Bavarian Academy of Sciences (LRZ). Gerald Mathias, Deputy Leader of the LRZ Application Group, explained that their interest in Spack emerged about two years ago when they were undergoing the procurement process for SuperMUC-NG, LRZ’s new 26.9 petaflops HPC system. On the predecessor system, SuperMUC, they managed the HPC software stack for users individually for each packet. Over the years, the stack had grown so large that they realized it wouldn’t be feasible to continue the practice with the larger system. That’s when they began to port their stack into Spack.

In late 2018, LRZ deployed their beta release of a Spack stack on their heterogeneous Linux cluster which serves local Munich universities. At the same time, they were working on the deployment and acceptance phase of SuperMUC-NG where it was the primary software stack at the start of user operations in August 2019. “Currently, our Spack-managed stack yields about 110 packages which supply some 220 modules to our users,” said Mathias. “Step-by-step, we continue to include more packages, particularly custom software, commercial software and core software-like compilers and MPI libraries,” he added.

Gerald Mathias, LRZ

For the future, they intend to push hard to keep pace with Spack’s rapid development. A key target is to provide a preconfigured Spack instance in user spaces. This will allow users to configure and install their own software, individually, but based on the common software stack provided by LRZ. “Furthermore, the definition of ‘environments’ seems very promising to compose easy-to-use software stacks for different scientific tasks and domains,” Mathias added.

While the majority of use cases represent X86 CPU/GPU environments, the Japanese Flagship Supercomputer Project being conducted by RIKEN, uses Spack to develop packages for their Arm (A64FX) system, Fugaku (formerly Post-K).

“Fugaku is optimized for large scale HPC and AI workflows; for this, together with our industry partner Fujitsu, we developed an HPC optimized Arm chip A64FX, that is three times as fast, and three times more power efficient than any other CPUs, as acknowledged by recent Green500 and other benchmarks. A64FX machines will be commercialized not only by Fujitsu but also Cray/HPE,” said RIKEN Center for Computational Science Director Satoshi Matsuoka (above photo, center). A64FX complies with Arm V8 and server standards; and Fugaku’s software stack is built on top of Red Hat Enterprise Linux. RIKEN and Fujitsu collaborate closely with the Spack development team to fix issues associated with aarch64; Arm 64-bit builds. Being one of the first exascale or near-exascale machines, both hardware and software must scale to an unprecedented magnitude of more than 150,000 nodes. Largely due to such effort centered around Spack, Matsuoka expects that everything running on current Arm and X86 clusters will ultimately run on Fugaku and other A64FX machines, but much, much faster.

According to Matsuoka, Fugaku’s first six racks will arrive at RIKEN this week (December 3, 2019), and by summer 2020, more than 400 racks will be in place. Early adopters will use the system this winter, and once performance benchmark are satisfied in December 2020, the system will go into full production in early 2021.

Glenn Johnson, UI

But not all Spack contributors work at large laboratories with massive systems; in fact, the largest number of contributors work with small to mid-sized clusters in academic research computing environments. HPC Architect Glenn Johnson (University of Iowa) was recognized as a feature contributor on Spack’s DOE R&D 100 Award (Software & Services). He is using Spack on Iowa’s Argon system that features 15,000 cores and 250 GPUs.

What motivates Johnson to contribute? He said, “Building and distributing scientific software applications is very challenging and becoming more so over time,” said Johnson. “Spack provides a tool, and a community, to develop best practices and standards for building and distributing scientific software, while still allowing for site variability. With Spack, as a tool, it is possible to benefit from the work of others while Spack, as a community, provides the opportunity to contribute back and realize the satisfaction that comes from having others benefit from your work,” he added.

As for Spack’s roadmap, DOE labs are leading the effort to detect and label installs for microarchitectures (Skylake, Zen, etc.). They’re optimizing packages for both cloud and static HPC environments and planning to focus on container integration; multi-mode, parallel builds; and better external package detection. They’re also developing a prototype for a new concretizer. In 2020, they will launch a multi-stage container builds that automatically exclude build dependencies from artifacts.

When asked if Spack is sustainable, Gamblin said that the DOE Advanced Simulation and Computing (ASC) program and Exascale Computing Project are funding a strong core team of developers, and are fully committed. The collaboration with Sylabs, which has a user base of 2.5 million developers worldwide, will help establish an even broader base of adoption.

A National Science Foundation/DOE tutorial will soon be added to their website: spack.io.

Photos by Leake, LLNL, and Fumikazu KONISHI. 

About the Author

HPCwire Contributing Editor Elizabeth Leake is a consultant, correspondent and advocate who serves the global high performance computing (HPC) and data science industries. In 2012, she founded STEM-Trek, a global, grassroots nonprofit organization that supports workforce development opportunities for science, technology, engineering and mathematics (STEM) scholars from underserved regions and underrepresented groups.

As a program director, Leake has mentored hundreds of early-career professionals who are breaking cultural barriers in an effort to accelerate scientific and engineering discoveries. Her multinational programs have specific themes that resonate with global stakeholders, such as food security data science, blockchain for social good, cybersecurity/risk mitigation, and more. As a conference blogger and communicator, her work drew recognition when STEM-Trek received the 2016 and 2017 HPCwire Editors’ Choice Awards for Workforce Diversity Leadership.

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!

Graphcore Introduces Larger-Than-Ever IPU-Based Pods

October 22, 2021

Graphcore and its “Intelligent Processing Units” (IPUs) emerged from stealth in 2016 and launched its second-generation IPU in 2020. While the company has also launched its IPUs in a variety of form factors over the Read more…

Quantum Chemistry Project to Be Among the First on EuroHPC’s LUMI System

October 22, 2021

Finland’s CSC has just installed the first module of LUMI, a 550-peak petaflops system supported by the European Union’s EuroHPC Joint Undertaking. While LUMI -- pictured in the header -- isn’t slated to complete i Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat for that approach), announced it was expanding into gate-based Read more…

LLNL Prepares the Water and Power Infrastructure for El Capitan

October 21, 2021

When it’s (ostensibly) ready in early 2023, El Capitan is expected to deliver in excess of two exaflops of peak computing power – around four times the power of Fugaku, the current top-ranked supercomputer in the wor Read more…

AWS Solution Channel

Royalty-free stock illustration ID: 537899029

Running GROMACS on GPU instances

Comparing the performance of real applications across different Amazon Elastic Compute Cloud (Amazon EC2) instance types is the best way we’ve found for finding optimal configurations for HPC applications here at AWS. Read more…

Faster Optical Switch that Operates at ‘Room Temp’ Developed by IBM, Skolkovo Researchers

October 19, 2021

Optical switching technology holds great promise for many applications but hot operating temperatures have been a key obstacle slowing progress. Now, a new optical switching device that can operate at room temperatures a Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

LLNL Prepares the Water and Power Infrastructure for El Capitan

October 21, 2021

When it’s (ostensibly) ready in early 2023, El Capitan is expected to deliver in excess of two exaflops of peak computing power – around four times the powe Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Quantum Workforce – NSTC Report Highlights Need for International Talent

October 13, 2021

Attracting and training the needed quantum workforce to fuel the ongoing quantum information sciences (QIS) revolution is a hot topic these days. Last week, the U.S. National Science and Technology Council issued a report – The Role of International Talent in Quantum Information Science... Read more…

Eni Returns to HPE for ‘HPC4’ Refresh via GreenLake

October 13, 2021

Italian energy company Eni is upgrading its HPC4 system with new gear from HPE that will be installed in Eni’s Green Data Center in Ferrera Erbognone (a provi Read more…

The Blueprint for the National Strategic Computing Reserve

October 12, 2021

Over the last year, the HPC community has been buzzing with the possibility of a National Strategic Computing Reserve (NSCR). An in-utero brainchild of the COVID-19 High-Performance Computing Consortium, an NSCR would serve as a Merchant Marine for urgent computing... Read more…

UCLA Researchers Report Largest Chiplet Design and Early Prototyping

October 12, 2021

What’s the best path forward for large-scale chip/system integration? Good question. Cerebras has set a high bar with its wafer scale engine 2 (WSE-2); it has 2.6 trillion transistors, including 850,000 cores, and was fabricated using TSMC’s 7nm process on a roughly 8” x 8” silicon footprint. Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

US Closes in on Exascale: Frontier Installation Is Underway

September 29, 2021

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, held by Zoom this week (Sept. 29-30), it was revealed that the Frontier supercomputer is currently being installed at Oak Ridge National Laboratory in Oak Ridge, Tenn. The staff at the Oak Ridge Leadership... Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer... Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was Read more…

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

Leading Solution Providers

Contributors

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

The Latest MLPerf Inference Results: Nvidia GPUs Hold Sway but Here Come CPUs and Intel

September 22, 2021

The latest round of MLPerf inference benchmark (v 1.1) results was released today and Nvidia again dominated, sweeping the top spots in the closed (apples-to-ap Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

Intel Unveils New Node Names; Sapphire Rapids Is Now an ‘Intel 7’ CPU

July 27, 2021

What's a preeminent chip company to do when its process node technology lags the competition by (roughly) one generation, but outmoded naming conventions make i Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

Quantum Computer Market Headed to $830M in 2024

September 13, 2021

What is one to make of the quantum computing market? Energized (lots of funding) but still chaotic and advancing in unpredictable ways (e.g. competing qubit tec Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire