Eitan Frachtenberg
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Eitan Frachtenberg has a total of 112 co-authors. Here are the top co-authors by number of shared publications:
- Dejan Milojicic
Hewlett Packard Enterprise, Milpitas, USA
15 shared publications F. Petrini
Applied Computer Science Group, Computational Sciences and Mathematics Division, Pacific Northwest National Laboratory, MS K7-90, Richland, WA 99352
13 shared publicationsJ. Fernandez
Departamento Ingenierı´a y Tecnologia de Computadores, Facultad de Informa´tica, Universidad de Murcia, 30071 Espinardo, Murcia, Spain
8 shared publications- Paolo Faraboschi
Artificial Intelligence Research Lab (AIRL), Hewlett Packard Labs, Milpitas, CA, USA
7 shared publications S. Coll
7 shared publications- Aditya Dhakal
Hewlett Packard Enterprise, Milpitas, USA
5 shared publications
Latest Publications
- 29 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 59 (6) , 26-39
The Industrial Revolution of the Intelligence Age
Humanity is entering an industrial revolution driven by intelligence, with artificial intelligence (AI) automating routine tasks at scale, transforming productivity and creativity, while elevating human attention and empathy as valuable resources. This article examines a future where society, markets, and human labor are reshaped by AI systems.
- 29 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 59 (6) , 26-39
The Industrial Revolution of the Intelligence Age
Humanity is entering an industrial revolution driven by intelligence, with artificial intelligence (AI) automating routine tasks at scale, transforming productivity and creativity, while elevating human attention and empathy as valuable resources. This article examines a future where society, markets, and human labor are reshaped by AI systems.
- Open Access
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 364-377
Are We There Yet? Predicting if Executing Applications are Near Completion
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 364-377
Open AccessAre We There Yet? Predicting if Executing Applications are Near Completion
- Open Access
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 277-286
OASIS: Optimal Allocation Strategy for Inference Services in Cloud Environments
Provisioning language models as inference services can be significantly challenging to cloud providers because of the need to balance service-level objectives (SLOs) with capital and operational costs. The heterogeneous nature of user queries, combined with temporal variations in workload patterns and the high energy consumption of GPUs warrants an intelligent resource-allocation strategy. We present OASIS (Optimal Allocation Strategy for Inference Services), a comprehensive methodology that combines static and dynamic optimizations to provision inference services efficiently. OASIS employs a two-phase approach: (1) pre-deployment profiling to determine optimal parameter configurations for different query types across various hardware options, and (2) run-time query routing and dynamic provisioning based on observed workload patterns. We evaluate OASIS using real-world BurstGPT traces on NVIDIA A100 and H100 GPUs. Our results show that: (1) GPU frequency scaling from 1320 MHz to 855 MHz reduces power by 39% with only 7% throughput loss; (2) query type classification reveals long-input-short-output workloads achieve 2.1× higher throughput than short-input-long-output; (3) MIG-based multi-tenancy achieves 42-43% power reduction when running two models simultaneously; and (4) runtime adaptation on real campus traces achieves statistically significant 12.4% power reduction while maintaining SLO compliance and identical throughput.
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 277-286
Open AccessOASIS: Optimal Allocation Strategy for Inference Services in Cloud Environments
Provisioning language models as inference services can be significantly challenging to cloud providers because of the need to balance service-level objectives (SLOs) with capital and operational costs. The heterogeneous nature of user queries, combined with temporal variations in workload patterns and the high energy consumption of GPUs warrants an intelligent resource-allocation strategy. We present OASIS (Optimal Allocation Strategy for Inference Services), a comprehensive methodology that combines static and dynamic optimizations to provision inference services efficiently. OASIS employs a two-phase approach: (1) pre-deployment profiling to determine optimal parameter configurations for different query types across various hardware options, and (2) run-time query routing and dynamic provisioning based on observed workload patterns. We evaluate OASIS using real-world BurstGPT traces on NVIDIA A100 and H100 GPUs. Our results show that: (1) GPU frequency scaling from 1320 MHz to 855 MHz reduces power by 39% with only 7% throughput loss; (2) query type classification reveals long-input-short-output workloads achieve 2.1× higher throughput than short-input-long-output; (3) MIG-based multi-tenancy achieves 42-43% power reduction when running two models simultaneously; and (4) runtime adaptation on real campus traces achieves statistically significant 12.4% power reduction while maintaining SLO compliance and identical throughput.
- Open Access
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 192-207
Variability-Guided Performance Optimization
The past few decades have seen software and hardware growing more heterogeneous and layered in abstractions. This trend produced many benefits for hiding complexity and increasing efficiency and modularity. But it also makes reasoning about performance and identifying its underlying factors more challenging because of the presence of performance variability. Moreover, performance variability can prevent synchronous applications from scaling and server applications from meeting service-level agreements. In this paper, we present a variability-guided optimization (VGO) workflow that leverages information in performance distributions to optimize performance, and more importantly, to reduce performance variability. It works by uncovering software, hardware, and compiler factors associated with specific aspects of variability and then suggesting measures for reducing it. Our experimental evaluation on a set of CPU and GPU benchmarks and applications shows that our tool successfully reduces the standard deviation and coefficient of variation of application run times by 0.374× and 0.444×, while also reducing mean run time by 0.843×. This technique enables tuning applications and their environments, improving their performance and predictability.
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 192-207
Open AccessVariability-Guided Performance Optimization
The past few decades have seen software and hardware growing more heterogeneous and layered in abstractions. This trend produced many benefits for hiding complexity and increasing efficiency and modularity. But it also makes reasoning about performance and identifying its underlying factors more challenging because of the presence of performance variability. Moreover, performance variability can prevent synchronous applications from scaling and server applications from meeting service-level agreements. In this paper, we present a variability-guided optimization (VGO) workflow that leverages information in performance distributions to optimize performance, and more importantly, to reduce performance variability. It works by uncovering software, hardware, and compiler factors associated with specific aspects of variability and then suggesting measures for reducing it. Our experimental evaluation on a set of CPU and GPU benchmarks and applications shows that our tool successfully reduces the standard deviation and coefficient of variation of application run times by 0.374× and 0.444×, while also reducing mean run time by 0.843×. This technique enables tuning applications and their environments, improving their performance and predictability.
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 59 (4) , 172-181
Technology Predictions 2026
IEEE Computer Society technology experts have unveiled 26 breakthrough technologies set to redefine industries and shape the future of our world for decades to come.
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 59 (4) , 172-181
Technology Predictions 2026
IEEE Computer Society technology experts have unveiled 26 breakthrough technologies set to redefine industries and shape the future of our world for decades to come.
- Open Access
- 5 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 320-333
HeteroBench: Multi-kernel Benchmarks for Heterogeneous Systems
- 5 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 320-333
Open AccessHeteroBench: Multi-kernel Benchmarks for Heterogeneous Systems
- 2 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 57 (4) , 143-149
The Distribution Is the Performance
In this column, we suggest how to sharpen our understanding of computer performance evaluation in light of variability and heterogeneity.
- 2 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 57 (4) , 143-149
The Distribution Is the Performance
In this column, we suggest how to sharpen our understanding of computer performance evaluation in light of variability and heterogeneity.
- 12 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IT Professional
- Vol. 26 (1) , 26-31
What Gets You Hired Now Will Not Get You Hired Then
Since its inception, humanity has depended on the skills of individuals and groups. Over millennia, humanity and skills evolved, differentiating those who would prosper from those who did not. From gatherers to hunters, from agriculture to industry, and from IT to AI, the complexity and rate of change have increased exponentially, driven by language, curiosity, and communication. IT skills and a variety of IT professions have dominated the past 25 years. At this juncture, advances in AI have altered the technology landscape, causing tectonic shifts in many professions. How will IT professions evolve, and how should IT professionals adapt? We contend that, first, required skill sets will rapidly change, increasing the importance of continuing education. Second, with the increased adoption of AI, the importance of data will also increase, demanding an ever-growing need for data science skills. Finally, many IT activities will be automated, requiring IT professionals to collaborate with AI assistants and take more strategic roles.
- 12 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IT Professional
- Vol. 26 (1) , 26-31
What Gets You Hired Now Will Not Get You Hired Then
Since its inception, humanity has depended on the skills of individuals and groups. Over millennia, humanity and skills evolved, differentiating those who would prosper from those who did not. From gatherers to hunters, from agriculture to industry, and from IT to AI, the complexity and rate of change have increased exponentially, driven by language, curiosity, and communication. IT skills and a variety of IT professions have dominated the past 25 years. At this juncture, advances in AI have altered the technology landscape, causing tectonic shifts in many professions. How will IT professions evolve, and how should IT professionals adapt? We contend that, first, required skill sets will rapidly change, increasing the importance of continuing education. Second, with the increased adoption of AI, the importance of data will also increase, demanding an ever-growing need for data science skills. Finally, many IT activities will be automated, requiring IT professionals to collaborate with AI assistants and take more strategic roles.
- 17 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 23 (1) , 21-24
Direct-Coding DNA With Multilevel Parallelism
The cost and time to sequence entire genomes have been on a steady and rapid decline since the early 2000 s, leading to an explosion of genomic data. In contrast, the growth rates for digital storage device capacity, CPU clock speed, and networking bandwidth have been much more moderate. This gap means that the need for storing, transmitting, and processing sequenced genomic data is outpacing the capacities of the underlying technologies. Compounding the problem is the fact that traditional data compression techniques used for natural language or images are not optimal for genomic data. To address this challenge, many data-compression techniques have been developed, offering a range of tradeoffs between compression ratio, computation time, memory requirements, and complexity. This paper focuses on a specific technique on one extreme of this tradeoff, namely two-bit coding, wherein every base in a genomic sequence is compressed from its original 8-bit ASCII representation to a unique two-bit binary representation. Even for this simple direct-coding scheme, current implementations leave room for significant performance improvements. Here, we show that this encoding can exploit multiple levels of parallelism in modern computer architectures to maximize encoding and decoding efficiency. Our open-source implementation achieves encoding and decoding rates of billions of bases per second, which are much higher than previously reported results. In fact, our measured throughput is typically limited only by the speed of the underlying storage media.
- 17 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 23 (1) , 21-24
Direct-Coding DNA With Multilevel Parallelism
The cost and time to sequence entire genomes have been on a steady and rapid decline since the early 2000 s, leading to an explosion of genomic data. In contrast, the growth rates for digital storage device capacity, CPU clock speed, and networking bandwidth have been much more moderate. This gap means that the need for storing, transmitting, and processing sequenced genomic data is outpacing the capacities of the underlying technologies. Compounding the problem is the fact that traditional data compression techniques used for natural language or images are not optimal for genomic data. To address this challenge, many data-compression techniques have been developed, offering a range of tradeoffs between compression ratio, computation time, memory requirements, and complexity. This paper focuses on a specific technique on one extreme of this tradeoff, namely two-bit coding, wherein every base in a genomic sequence is compressed from its original 8-bit ASCII representation to a unique two-bit binary representation. Even for this simple direct-coding scheme, current implementations leave room for significant performance improvements. Here, we show that this encoding can exploit multiple levels of parallelism in modern computer architectures to maximize encoding and decoding efficiency. Our open-source implementation achieves encoding and decoding rates of billions of bases per second, which are much higher than previously reported results. In fact, our measured throughput is typically limited only by the speed of the underlying storage media.
- 3 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 57 (1) , 136-144
Predicting Heterogeneity and Serverless Principles of Converged High-Performance Computing, Artificial Intelligence, and Workflows
Traditional high-performance computing and modern artificial intelligence computing are converging with workflows as a common paradigm. We predict nine principles of heterogeneity and serverless computing for this convergence, from high-level programming to low-level hardware.
- 3 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 57 (1) , 136-144
Predicting Heterogeneity and Serverless Principles of Converged High-Performance Computing, Artificial Intelligence, and Workflows
Traditional high-performance computing and modern artificial intelligence computing are converging with workflows as a common paradigm. We predict nine principles of heterogeneity and serverless computing for this convergence, from high-level programming to low-level hardware.