Dejan Milojicic

Dejan Milojicic
HP Inc. | HP · Hewlett Packard Labs

Doctor of Engineering

About

235
Publications
57,062
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
6,260
Citations
Citations since 2017
51 Research Items
2349 Citations
20172018201920202021202220230100200300400500
20172018201920202021202220230100200300400500
20172018201920202021202220230100200300400500
20172018201920202021202220230100200300400500
Introduction

Publications

Publications (235)
Article
Extreme times require extreme measures. In this column, we discuss how high-performance computing embraces artificial intelligence and data analytics to address global challenges.
Preprint
Full-text available
Operating Systems are built upon a set of abstractions to provide resource management and programming APIs for common functionality, such as synchronization, communication, protection, and I/O. The process abstraction is the bridge across these two aspects; unsurprisingly, research efforts pay particular attention to the process abstraction, aiming...
Article
Photonics has been used for computer communication at larger distances. Due to performance, power, and packaging, photonics holds promise for moving further to and onto chips. This is called silicon photonics.
Article
Driven by artificial intelligence (AI), accelerators are taking away the “central” aspect of CPUs to become dominant processors of the vast amounts of generated data today. This transition happened in your phones and embedded devices a decade ago and, because of AI, is now taking place in servers and in the cloud.
Article
In this special issue, we have accepted five papers after rigorous reviews. They are as follows. 1) Smart home integration with external IoT device platforms and services. 2) Decentralized video input authentication as an edge service for smart cities. 3) Smart water-meter wireless transmission system for smart cities. 4) WATERSensing: A smart warn...
Article
Twenty-five years ago, no one expected services to be seamlessly and consistently delivered across campuses and edge. It is now possible to envision machine learning-powered services capable of scaling to the reliability, performance, and security requirements of billions of users worldwide.
Article
Driven by convergence with artificial intelligence and data analytics, increased heterogeneity, and a hybrid cloud/on-premise delivery model, dynamic composition of workflows will be a key design criteria of future high-performance computing (HPC) systems. While tightly coupled HPC workloads will continue to execute on dedicated supercomputers, oth...
Article
COVID-19 has accelerated and scaled out remote education beyond our wildest imagination. How can we retain benefits, eliminate downsides, and build education better for the long-term future?
Article
At a time of global uncertainty, this theme issue of Computer explores developments that could benefit humanity and the planet.
Preprint
Cloud deployments disaggregate storage from compute, providing more flexibility to both the storage and compute layers. In this paper, we explore disaggregation by taking it one step further and applying it to memory (DRAM). Disaggregated memory uses network attached DRAM as a way to decouple memory from CPU. In the context of databases, such a des...
Article
Cache coherence in modern computer architectures enables easier programming by sharing data across multiple processors. Unfortunately, it can also limit scalability due to cache coherency traffic initiated by competing memory accesses. Rack‐scale systems introduce shared memory across a whole rack, but without inter‐node cache coherence. This poses...
Article
Computer hosts a virtual roundtable with four past IEEE Computer Society presidents to discuss the original rationale for creating special technical communities and their future.
Article
Computer hosts a virtual roundtable with three experts to discuss the opportunities and obstacles regarding edge-to-cloud technology.
Article
Artificial intelligence will be deployed increasingly in more systems that affect public health, safety, and welfare. These systems will better utilize scarce resources; prevent disasters; and increase safety, reliability, comfort, and convenience. Despite the technological challenges and public fears, these systems will improve the quality of life...
Article
The wide adoption of deep neural networks has been accompanied by ever-increasing energy and performance demands due to the expensive nature of training them. Numerous special-purpose architectures have been proposed to accelerate training: both digital and hybrid digital-analog using resistive RAM (ReRAM) crossbars. ReRAM-based accelerators have d...
Article
This virtual roundtable discussion examines accelerators for artificial intelligence (AI) and high-performance computing (HPC).
Preprint
The wide adoption of deep neural networks has been accompanied by ever-increasing energy and performance demands due to the expensive nature of training them. Numerous special-purpose architectures have been proposed to accelerate training: both digital and hybrid digital-analog using resistive RAM (ReRAM) crossbars. ReRAM-based accelerators have d...
Article
Presents Computer Society's technology predictions. CS predictions have an audience of millions of readers, compared to the thousands who read traditional technical papers. This is partially because of the new content delivery models, but it is also due to the lighter weight and shorter format. Scorecards get even more attention: the audience wants...
Article
Making accurate predictions isn't easy. However, many people have enjoyed making predictions throughout history, and even more individuals have taken part in learning about predictions.
Article
Cybersecurity is a key risk for any business as the number of attacks are increasing. Growing attacks on cybersecurity are threatening our existence. Artificial Intelligence (AI) and machine learning (ML) can help detect threats and provide recommendations to cyber analysts. Advancing the adoption of AI/ML applied to cybersecurity requires partneri...
Preprint
In this paper, we rethink how an OS supports virtual memory. Classical VM is an opaque abstraction of RAM, backed by demand paging. However, most systems today (from phones to data-centers) do not page, and indeed may require the performance benefits of non-paged physical memory, precise NUMA allocation, etc. Moreover, MMU hardware is now useful fo...
Article
Full-text available
DevOpsis a collaborative and multidisciplinary organizational effort to automate continuous delivery of new software updates while guaranteeing their correctness and reliability. The present survey investigates and discusses DevOps challenges from the perspective of engineers, managers, and researchers. We review the literature and develop a DevOps...
Conference Paper
Server systems with large amounts of physical memory can benefit from using some of the available memory capacity for in-memory snapshots of the ongoing computations. In-memory snapshots are useful for services such as scaling of new workload instances, debugging, during scheduling, etc., which do not require snapshot persistence across node crashe...
Preprint
Full-text available
DevOps is a collaborative and multidisciplinary organizational effort to automate continuous delivery of new software updates while guaranteeing their correctness and reliability. The present survey investigates and discusses DevOps challenges from the perspective of engineers, managers, and researchers. We review the literature and develop a DevOp...
Conference Paper
Full-text available
Memristor crossbars are circuits capable of performing analog matrix-vector multiplications, overcoming the fundamental energy efficiency limitations of digital logic. They have been shown to be effective in special-purpose accelerators for a limited set of neural network applications. We present the Programmable Ultra-efficient Memristor-based Acc...
Conference Paper
Heterogeneous CPU-FPGA systems are evolving towards tighter integration between CPUs and FPGAs for improved performance and energy efficiency. At the same time, programmability is also improving with High Level Synthesis tools (e.g., OpenCL Software Development Kits), which allow programmers to express their designs with high-level programming lang...
Article
Byte-addressable nonvolatile memory (NVM) blends the concepts of storage and memory and can radically improve data-centric applications, from in-memory databases to graph processing. By enabling large-capacity devices to be shared across multiple computing elements, fabric-attached NVM changes the nature of rack-scale systems and enables short-late...
Article
Full-text available
For decades, computer architectures have treated memory and storage as separate entities. Today we watch the emergence of new memory technologies that promise to significantly change the landscape of memory systems. Byte-addressable Non-Volatile Memory (NVM) technologies are expected to offer access latency close to that of Dynamic Random Access Me...
Article
Rack-scale systems with large, shared, disaggregated, and persistent memory need solid protection and authorization techniques. Our solution uses a memoryside capability enforcement processor that gates memory accesses through extended capabilities, enables finegrained access control beyond a single address space, and minimally disrupts the program...
Preprint
Full-text available
Memristor crossbars are circuits capable of performing analog matrix-vector multiplications, overcoming the fundamental energy efficiency limitations of digital logic. They have been shown to be effective in special-purpose accelerators for a limited set of neural network applications. We present the Programmable Ultra-efficient Memristor-based Acc...
Article
Full-text available
The Cloud computing paradigm has revolutionized the computer science horizon during the past decade and has enabled the emergence of computing as the fifth utility. It has captured significant attention of academia, industries and government bodies. Now, it has emerged as the backbone of modern economy by offering subscription-based services anytim...
Chapter
Full-text available
Large cities around the world face numerous challenges to guarantee the quality of life of its citizens. A promising approach to cope with these problems is the concept of Smart Cities, of which the main idea is the use of Information and Communication Technologies to improve city services. Being able to simulate the execution of Smart Cities scena...
Article
Security vulnerabilities such as Meltdown and Spectre demonstrate how chip complexity grew faster than our ability to manage unintended consequences. Attention to security from the outset should be part of the rememdy, yet complexity must be controlled at a more fundamental level.
Article
Key-value stores are an important tool in managing and accessing large in-memory data sets. As many applications benefit from having as much of their working state fit into main memory, an important design of the memory management of modern key-value stores is the use of log-structured approaches, enabling efficient use of the memory capacity, by c...
Article
Direct sharing and storing of memory objects allows high-performance and low-overhead collaboration between parallel processes or application workflows with loosely coupled programs. However, sharing of objects is hindered by the inability to use subtype polymorphism which is common in object-oriented programming languages. That is because implemen...
Conference Paper
Full-text available
Large cities around the world face numerous challenges to guarantee the quality of life of its citizens. A promising approach to cope with these problems is the concept of Smart Cities, of which the main idea is the use of Information and Communication Technologies to improve city services. Being able to simulate the execution of Smart Cities scena...
Conference Paper
It is time to reconsider memory protection. The emergence of large non-volatile main memories, scalable interconnects, and rack-scale computers running large numbers of small "micro services" creates significant challenges for memory protection based solely on MMU mechanisms. Central to this is a tension between protection and translation: optimizi...
Conference Paper
as the deceleration of processor scaling due to Moore's law accelerates research in new types of computing structures, the need arises for rethinking operating systems paradigms. Traditionally, an operating system is a layer between hardware and applications and its primary function is in managing hardware resources and providing a common abstracti...
Article
Full-text available
Making cities smarter help improve city services and increase citizens' quality of life. Information and communication technologies (ICT) are fundamental for progressing towards smarter city environments. Smart City software platforms potentially support the development and integration of Smart City applications. However, the ICT community must ove...
Conference Paper
Full-text available
Smart Cities, i.e., cities enhanced with a technological infrastructure that enables a more intelligent use and management of its resources, are currently seen as a powerful way of improving the quality of life of its citizens. Smart city platforms tailored at metropolises will be intrinsically very-large-scale systems; designing and developing suc...
Article
Full-text available
In ever-changing business environments, organizations continuously refine their processes to benefit from and meet the constraints of new technology, new business rules, and new market requirements. Workflow management systems (WFMSs) support organizations in evolving their processes by providing them with technological mechanisms to design, enact,...
Article
Memory-centric computing demands careful organization of the virtual address space, but traditional methods for doing so are inflexible and inefficient. If an application wishes to address larger physical memory than virtual address bits allow, if it wishes to maintain pointer-based data structures beyond process lifetimes, or if it wishes to share...
Article
Memory-centric computing demands careful organization of the virtual address space, but traditional methods for doing so are inflexible and inefficient. If an application wishes to address larger physical memory than virtual address bits allow, if it wishes to maintain pointer-based data structures beyond process lifetimes, or if it wishes to share...
Article
Memory-centric computing demands careful organization of the virtual address space, but traditional methods for doing so are inflexible and inefficient. If an application wishes to address larger physical memory than virtual address bits allow, if it wishes to maintain pointer-based data structures beyond process lifetimes, or if it wishes to share...
Conference Paper
Memory-centric computing demands careful organization of the virtual address space, but traditional methods for doing so are inflexible and inefficient. If an application wishes to address larger physical memory than virtual address bits allow, if it wishes to maintain pointer-based data structures beyond process lifetimes, or if it wishes to share...
Article
Will OSs in 2025 still resemble the Unix-like consensus of today, or will a very different design achieve widespread adoption?
Conference Paper
Modern data center clusters are shifting from dedicated single framework clusters to shared clusters. In such shared environments, cluster schedulers typically utilize preemption by simply killing jobs in order to achieve resource priority and fairness during peak utilization. This can cause significant resource waste and delay job response time. I...
Article
Over the decades, standards have been critical for defining how to interconnect computer and networking devices across different vendors so they can seamlessly work together. Standards have been critical, not only in networking and computer interfaces, but also at the operating system and systems software level. There are many examples, such as IEE...
Conference Paper
By the end of the decade, computing designs will shift from a processor-centric architecture to a memory-centric architecture. At rack scale, we can expect a large pool of non-volatile memory (NVM) that will be accessed by heterogeneous and decentralized compute resources [3, 17]. Such memory-centric architectures will present challenges that today...
Article
Full-text available
In creating the CS 2022 Report, we were searching for a meta-innovation that would tie all these technology areas together. We found a unifying theme in seamless intelligence, where everything is connected through ubiquitous networks, interfaces, and so on. While similar to previous pervasive and ubiquitous computing scenarios, seamless intelligenc...
Patent
Assigning nodes to jobs based on reliability factors includes calculating the maximum value of a processor utilization efficiency and assigning an optimal number of spare nodes to the job based on the value of the processor utilization efficiency.
Article
The Computer Society's revitalization was an effort driven by its entire leadership team and staff--we shifted focus (including volunteers, staff, and finances) from less productive areas and began to seriously invest in new ones.
Patent
A technique includes using a computer agent to observe diagnoses of computer-related incidents. Based on the observation, patterns are identified in the diagnoses, and based at least in part on the patterns, the diagnoses are selectively automated.
Article
Full-text available
Cloud computing is emerging as a promising alternative to supercomputers for some high-performance computing (HPC) applications. With cloud as an additional deployment option, HPC users and providers are faced with the challenges of dealing with highly heterogeneous resources, where the variability spans across a wide range of processor configurati...
Article
Editor-in-Chief Misha Rabinovich interviews Dejan Milojicic, IEEE Computer Society president, and Vint Cerf, president of ACM, on various topics of interest to IC's readers, including net neutrality, the Future Internet, and the next big app. The full interview transcript is available as a Web Extra accompanying this article. The full interview can...
Patent
Service level objectives for a multi-tier system are translated to low-level system metrics by determining component profiles. The component profiles include performance parameters calculated as a function of applied resources for each of a plurality of components of the multi-tier system. Performance models including service level parameters are a...
Article
Full-text available
In this paper, we explore the automation of services' compositions. We focus on the service selection problem. In the formulation that we consider, the problem's inputs are constituted by a behavioral composition whose abstract services must be bound to concrete ones. The objective is to find the binding that optimizes the {\it utility} of the comp...
Article
Today's IT professionals, computer engineers, and students share knowledge directly through the Internet and gather at cool events like hackathons. How can the Computer Society reinvent its relevance with contemporary researchers and practitioners?
Article
Choreographies are a distributed approach for composing web services. Compared to orchestrations, which use a centralized scheme for distributed service management, the interaction among the choreographed services is collaborative with decentralized coordination. Despite the advantages, choreography development, including the testing activities, ha...
Conference Paper
Full-text available
Cloud computing is emerging as an alternative to supercomputers for some of the high-performance computing (HPC) applications that do not require a fully dedicated machine. With cloud as an additional deployment option, HPC users are faced with the challenges of dealing with highly heterogeneous resources, where the variability spans across a wide...
Article
Members also elect new vice presidents and Board of Governors members.
Patent
A subtree within a global tree of nodes is created by determining a predicate condition. The predicate condition is disseminated to the nodes in the global tree. For each node in the global tree, a determination of whether the node belongs to the subtree is performed, and an indication of whether the node belongs to the subtree is stored. After the...
Conference Paper
Full-text available
Large and complex workflow repositories include a series of interdependent workflows. In this scenario, it becomes hard to estimate the effort required to accomplish changes to workflows. Furthermore, ad-hoc changes may induce side and ripple effects, which ultimately hamper the reliability of the repository. In this paper, we introduce a static de...
Article
Full-text available
The articles in this special section focus on the topic of cloud computing, technologies, applications, and new areas of technological innovation.
Conference Paper
Full-text available
Driven by the benefits of elasticity and pay-as-you-go model, cloud computing is emerging as an attractive alternative and addition to in-house clusters and supercomputers for some High Performance Computing (HPC) applications. However, poor interconnect performance, heterogeneous and dynamic environment, and interference by other virtual machines...
Conference Paper
Full-text available
Rapid checkpointing will remain key functionality for next generation high end machines. This paper explores the use of node-local nonvolatile memories (NVM) such as phase-change memory, to provide frequent, low overhead checkpoints. By adapting existing multi-level checkpoint techniques, we devise new methods, termed NVM-checkpoints, that efficien...
Conference Paper
Full-text available
Cloud offerings are increasingly serving workloads with a large variability in terms of compute, storage and net- working resources. Computing requirements (all the way to High Performance Computing or HPC), criticality, communication intensity, memory requirements, and scale can vary widely. Virtual Machine (VM) placement and consolidation for eff...
Article
Full-text available
A challenging task in Web service composition is the runtime binding of a set of interconnected abstract services to concrete ones. This question, formulated as the service selection problem, has been studied in the area of service compositions implementing business processes. Despite the abundance of work on this topic, few of them match some prac...
Conference Paper
Full-text available
Given a Web Services Composition, we deal with the prediction of the mean service response time that can be expected from a user request that is serviced. This challenge is a key issue in the design of middleware, managing Web Services Composition. We focus on complex services composition that can be described as BPMN choreographies of services. Ou...