Mahadev Satyanarayanan

Carnegie Mellon University | CMU · Computer Science Department

Virtual machine (VM) migration demands distinct properties under resource oversubscription and workload surges. We present enlightened post-copy, a new mechanism for VMs under contention that evicts the target VM with fast execution transfer and short total duration. This design contrasts with common live migration, which uses the down time of the...

Computational offloading services at the edge of the Internet for mobile devices are becoming a reality. Using a wide range of mobile applications, we explore how such infrastructure improves latency and energy consumption relative to the cloud. We present experimental results from WiFi and 4G LTE networks that confirm substantial wins from edge co...

Virtual machine (VM) migration demands distinct properties under resource oversubscription and workload surges. We present enlightened post-copy, a new mechanism for VMs under contention that evicts the target VM with fast execution transfer and short total duration. This design contrasts with common live migration, which uses the down time of the...

Unease over data privacy will retard consumer acceptance of IoT deployments. The primary source of discomfort is a lack of user control over raw data that is streamed directly from sensors to the cloud. This is a direct consequence of the over-centralization of today's cloud-based IoT hub designs. We propose a solution that interposes a locally-con...

Tech report of paper in review

Mahadev Satyanarayanan (Satya) presented his thoughts on "The Evolution of Memory and File Systems". He observed that over a 60-year period, there have been four drivers of progress: the quests for scale, performance, transparency, and robustness. At the dawn of computing, the quest for scale was dominant. Easing the memory limitations of early com...

Background: Computer-assisted diagnosis of dermoscopic images of skin lesions has the potential to improve melanoma early detection. Objective: We sought to evaluate the performance of a novel classifier that uses decision forest classification of dermoscopic images to generate a lesion severity score. Methods: Severity scores were calculated...

A cognitive assistance application combines a wearable device such as Google Glass with cloudlet processing to provide step-by-step guidance on a complex task. In this paper, we focus on user assistance for narrow and well-defined tasks that require specialized knowledge and/or skills. We describe proof-of-concept implementations for four different...

High-data-rate sensors, such as video cameras, are becoming ubiquitous in the Internet of Things. This article describes GigaSight, an Internet-scale repository of crowd-sourced video content, with strong enforcement of privacy preferences and access controls. The GigaSight architecture is a federated system of VM-based cloudlets that perform video...

Modern offices are crowded with personal computers. While studies have shown these to be idle most of the time, they remain powered, consuming up to 60% of their peak power. Hardware-based solutions engendered by PC vendors (e.g., low-power states, Wake-on-LAN) have proved unsuccessful because, in spite of user inactivity, these machines often need...

We show how a disruptive force in mobile computing can be created by extending today???s unmodified cloud to a second level consisting of self-managed data centers with no hard state called cloudlets. These are located at the edge of the Internet, just one wireless hop away from associated mobile devices. By leveraging lowlatency offload, cloudlets...

When offloading computation from a mobile device, we show that it can pay to perform additional on-device work in order to reduce the offloading workload. We call this offload shap- ing, and demonstrate its application at many different levels of abstraction using a variety of techniques. We show that offload shaping can produce significant reducti...

As mobile computing and cloud computing converge, the sensing and interaction capabilities of mobile devices can be seamlessly fused with compute-intensive and data-intensive processing in the cloud. Cloudlets are important architectural components in this convergence, representing the middle tier of a mobile device - cloudlet - cloud hierarchy. We...

This is the final report on reproducibility@xsede, a one-day workshop held in conjunction with XSEDE14, the annual conference of the Extreme Science and Engineering Discovery Environment (XSEDE). The workshop's discussion-oriented agenda focused on reproducibility in large-scale computational research. Two important themes capture the spirit of the...

We describe the architecture and prototype implementation of an assistive system based on Google Glass devices for users in cognitive decline. It combines the first-person image capture and sensing capabilities of Glass with remote processing to perform real-time scene interpretation. The system architecture is multi-tiered. It offers tight end-to-...

Executable content is of growing importance in many domains. How does one share and archive such content at Internet-scale for spatial and temporal collaboration? Spatial collaboration refers to the classic concept of user collaboration: two or more users who are at different Internet locations performing a task using shared context. Temporal colla...

We propose a non-intrusive approach for moni-toring virtual machines (VMs) in the cloud. At the core of this approach is a mechanism for selective real-time monitoring of guest file updates within VM instances. This mechanism is agentless, requiring no guest VM support. It has low virtual I/O overhead, low latency for emitting file updates, and a s...

Effortless one-touch capture of video is a unique capability of wearable devices such as Google Glass. We use this capability to create a new type of crowd-sourced system in which users receive queries relevant to their current location and opt-in preferences. In response, they can send back live video snippets of their surroundings. A system of re...

The convergence of mobile computing and cloud computing is predicated on a reliable, high-bandwidth end-to-end network. This basic requirement is hard to guarantee in hostile environments such as military operations and disaster recovery. In this article, the authors examine how VM-based cloudlets that are located in close proximity to associated m...

Cloud-sourced virtual appliances (VAs) have been touted as powerful solutions for many software maintenance, mobility, backward compatibility, and security challenges. In this paper, we ask whether it is possible to create a VA cloud service that supports fluid, interactive user experience even over mobile networks. More specifically, we wish to su...

Although widely touted as a replacement for glass slides and microscopes in pathology, digital slides present major challenges in data storage, transmission, processing and interoperability. Since no universal data format is in widespread use for these images today, each vendor defines its own proprietary data formats, analysis tools, viewers and s...

We propose a scalable Internet system for continuous collection of crowd-sourced video from devices such as Google Glass. Our hybrid cloud architecture, GigaSight, is effectively a Content Delivery Network (CDN) in reverse. It achieves scalability by decentralizing the collection infrastructure using cloudlets based on virtual machines~(VMs). Based...

Cloud offload is an important technique in mobile computing. VM-based cloudlets have been proposed as offload sites for the resource-intensive and latency-sensitive computations typically associated with mobile multimedia applications. Since cloud offload relies on precisely-configured back-end software, it is difficult to support at global scale a...

The convergence of mobile computing and cloud computing assumes that the cloud is easily accessible at all times. In other words, there is good end-to-end network quality and few network or cloud failures. Unfortunately, this assumption is untenable in several important contexts that we collectively refer to as hostile environments. The prime examp...

Since the dawn of mobile computing two decades ago, the unique constraints of mobility have shaped the software architectures of systems. We now stand at the threshold of the next major transformation in computing: one in which the rich sensing and interaction capabilities of mobile devices are seamlessly fused with compute-intensive and data-inten...

This article investigates the transient use of free local storage for improving performance in VM-based mobile computing systems implemented as thick clients on host PCs. We use the term TransientPC systems to refer to these types of systems. The solution we propose, called TransPart, uses the higher-performing local storage of host hardware to spe...

The convergence of mobile computing and cloud computing enables new multimedia applications that are both resource-intensive and interaction-intensive. For these applications, end-to-end network bandwidth and latency matter greatly when cloud resources are used to augment the computational power and battery life of a mobile device. We first present...

Mobile crowdsensing is becoming a vital technique for environment monitoring, infrastructure management, and social computing. However, deploying mobile crowdsensing applications in large-scale environments is not a trivial task. It creates a tremendous burden on application developers as well as mobile users. In this paper we try to reveal the bar...

We present a cloud-based approach to opportunistic, near real-time search of untagged images on smartphones that is sensitive to bandwidth and energy constraints. Our approach is inspired by the long-established practice of photographers using contact sheets to rapidly visualize a new collection of photographs, and then selecting a subset on which...

Handheld mobile technology can help disaster relief workers and soldiers in the field with tasks such as speech and image recognition, natural language processing, decision-making, and mission planning. However, these applications are computation-intensive, take a heavy toll on battery power, and often rely on good connectivity to networks, limitin...

The capabilities of mobile devices have been improving very quickly in terms of computing power, storage, feature support, and developed applications. However, these mobile applications are still intrinsically limited by a relative lack of bandwidth, computing power, and energy compared to their tethered counterparts. Cloud computing offers abundan...

Handheld computing has had many applications in medicine, but relatively few in pathology. Most reported uses of handhelds in pathology have been limited to experimental endeavors in telemedicine or education. With recent advances in handheld hardware and software, along with concurrent advances in whole-slide imaging (WSI), new opportunities and c...

Idle desktop systems are frequently left powered, often because of applications that maintain network presence or to enable potential remote access. Unfortunately, an idle PC consumes up to 60% of its peak power. Solutions have been proposed that perform consolidation of idle desktop virtual machines. However, desktop VMs are often large requiring...

Transient use of displays by mobile users was prophesied two decades ago. Today, convergence of a range of technologies enable the realization of this vision. For researchers in this space, one key question is where to physically locate the application for which the display has been appropriated. The emergence of cloud and cloudlet computing has in...

The convergence of cloud computing with mobile computing opens the door to the creation of new applications and services that can be delivered to users at any time and any place. At the heart of this convergence lies a delicate balance between centralization and decentralization. We explore the forces underlying this balance, and examine the role o...

In the inaugural issue of MC2R in April 1997 [24], I highlighted the seminal influence of mobility in computing. At that time, the goal of "information at your fingertips anywhere, anytime" was only a dream. Today, through relentless pursuit of innovations in wireless technology, energy-efficient portable hardware and adaptive software, we have lar...

Photos taken by smartphone users can accidentally contain content that is timely and valuable to others, often in real-time. We report the system design and evaluation of a distributed search system, Theia, for crowd-sourced real-time content search of smartphone photos. Because smartphones are resource-constrained, Theia incorporates two key innov...

The success of cloud computing leads to large, centralized collections of virtual machine (VM) images. The ability to retrospect (examine the historical state of) these images at a high semantic level can be valuable in many aspects of IT management such as debugging and troubleshooting, software quality control, legal establishment of data or code...

The proliferation of pictures and videos in the Internet is imposing heavy demands on mobile data networks. This demand is expected to grow rapidly and a one-fit-all solution is unforeseeable. While researchers are approaching the problem from different directions, we identify a human-centric opportunity to reduce content size. Our intuition is tha...

A basic building block of cloud computing is virtualization. Virtual machines (VMs) encapsulate a user’s computing environment and efficiently isolate it from that of other users. VMs, however, are large entities, and no clear APIs exist yet to provide users with programatic, fine-grained control on short time scales. We present SnowFlock, a paradi...

"Information at your fingertips anywhere, anytime" has been the driving vision of mobile computing for the past two decades. Through relentless pursuit of this vision, spurring innovations in wireless technology, energy-efficient portable hardware and adaptive software, we have now largely attained this goal. Ubiquitous email and Web access is a re...

A basic building block of cloud computing is virtualization. Virtual machines (VMs) encapsulate a user’s computing environment and efficiently isolate it from that of other users. VMs, however, are large entities, and no clear APIs exist yet to provide users with programatic, fine-grained control on short time scales. We present SnowFlock, a paradi...

Executable content is of growing importance in many domains. How does one share and archive such content at Internet-scale for spatial and temporal collaboration? Spatial collaboration refers to the classic concept of user collaboration: two or more users who are at different Internet locations performing a task using shared context. Temporal colla...

Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily represent the views of the NSF, IBM, Rutgers University, or Carnegie Mellon University. Keywords: Internet Suspend/Resume, ISR, OpenISR, virtual machines, This paper investigates the transient use of free local storage f...

The success of cloud computing can lead to large, centralized collections of virtual machine (VM) images. The ability to interactively search these VM images at a high semantic level emerges as an important capability. This paper examines the opportunities and challenges in creating such a search capability, and presents early evidence of its feasi...

Office and home environments are increasingly crowded with personal computers. Even though these computers see little use in the course of the day, they often remain powered, even when idle. Leaving idle PCs running is not only wasteful, but with rising energy costs it is in-creasingly more expensive. We propose partial migration of idle desktop se...

Discard-based search is a new approach to searching the content of complex, unlabeled, nonindexed data such as digital photographs, medical images, and real-time surveillance data. The essence of this approach is query-specific content-based computation, pipelined with human cognition. In this approach, query-specific parallel computation shrinks a...

Similarity measurement is a critical component in content-based image retrieval systems, and learning a good distance metric can significantly improve retrieval performance. However, despite extensive study, there are several major shortcomings with the existing approaches for distance metric learning that can significantly affect their application...

Mobile computing continuously evolve through the sustained effort of many researchers. It seamlessly augments users' cognitive abilities via compute-intensive capabilities such as speech recognition, natural language processing, etc. By thus empowering mobile users, we could transform many areas of human activity. This article discusses the technic...

Mobility footprint refers to the size, weight, and energy de- mand of the hardware that must be carried by a mobile user to be eective at any time and place. The ideal of a zero mobility footprint is achievable by encapsulating per- sonal computing state in a virtual machine (VM) and de- livering it over the Internet to a locally-obtained computer...

Virtual Machine (VM) fork is a new cloud computing ab- straction that instantaneously clones a VM into multiple replicas running on different hosts. All replicas share the same initial state, matching the intuitive semantics of stateful worker creation. VM fork thus enables the straightforward creation and efficient deployment of many tasks demand-...

This paper presents a framework for engineering resource-adaptive software targeted at small mobile devices. Rather than building a solution from scratch, we extend and integrate existing work on software infrastructures for ubiquitous computing, and on resource-adaptive applications. This paper addresses two research questions: first, is it feasib...

Kimberley enables rapid software provisioning of fixed infrastructure for transient use by a mobile device. It uses virtual machine (VM) technology, but avoids the performance challenges of running VMs on resource-poor mobile devices. VM execution only occurs in the infrastructure and never on the mobile device; however, the device may transport an...

This paper presents an automated, online approach to anomaly detection in high-content screening assays for pharmaceutical research. Online detection of anomalies is attractive because it offers the possibility of immediate corrective action, early termination, and redesign of assays that may require many hours or days to execute. The proposed appr...

This paper presents a framework for engineering resource-adaptive software systems targeted at small mobile devices. The proposed framework empowers users to control tradeoffs among a rich set of service- specific aspects of quality of service. After motivating the problem, the paper proposes a model for capturing user preferences with respect to q...

Virtual machine (VM) migration has been proposed as a building block for mobile computing. An important challenge for VM migration is to optimize the transfer of large amounts of disk and memory state. We propose a solution based on opportunistic replay of user actions. In contrast to other replay techniques, opportunistic replay captures user inte...

Building an optimal image reference library is a critical step in developing the interactive computer-aided detection and diagnosis (I-CAD) systems of medical images using content-based image retrieval (CBIR) schemes. In this study, the authors conducted two experiments to investigate (1) the relationship between I-CAD performance and size of refer...

Although widely touted as a replacement for glass slides and microscopes, whole-slide images used in digital pathology present a challenge in analysis and interoperability. No universal data format exists for these images: analysis tools, viewers, and libraries are vendor-specific. In this paper, we present a vendor-neutral C library for reading wh...

Snowbird is a middleware system based on virtual machine (VM) technology that simplifies the development and deployment of bimodal applications. Such applications alternate between phases with heavy computational-resource needs and phases rich in user interaction. Examples include digital animation, as well as scientific, medical, and engineering d...

In the emerging Internet Suspend/Resume mobile computing model, users exploit pervasive deployments of inexpensive, mass- market PC hardware rather than carrying portable hardware. ISR's driving vision is that plummeting hardware costs will someday eliminate the need to carry computing environments in portable computers. Instead, ISR will deliver a...

Security and privacy concerns touch on all aspects of pervasive computing, including hardware, operating systems, networks, databases, user interfaces, and applications. The seven articles selected for this special issue draw on ideas from many of these fields and provide a flavor of the kinds of security and privacy challenges and opportunities in...

Security researchers and practitioners lack techniques to quantitatively evaluate the strength of security systems against a determined attacker. Currently, evaluation is either qualitative, such as through security certification standards, or ad-hoc, such as through penetration testing and auditing. In this paper, we propose a framework that if ap...

Self-adapting systems based on multiple concurrent applications must decide how to allocate scarce resources to applications and how to set the quality parameters of each application to best satisfy the user. Past work has made those decisions with analytic models that used current resource availability information: they react to recent changes in...

In the field of lipid research, the measurement of adipocyte size is an important but difficult problem. We describe an imaging-based solution that combines precise investigator control with semi-automated quantitation. By using unfixed live cells, we avoid many complications that arise in trying to isolate individual adipocytes. Instead, we image...