This paper addresses algorithms for dynamically varying (scaling) CPU speed and voltage in order to save energy. Such scaling is useful and effective when it is immaterial when a task completes, as long as it meets some deadline. We show how to modify any scaling algorithm to keep performance the same but minimize expected energy consumption. We refer to our approach as PACE (Processor Acceleration to Conserve Energy) since the resulting schedule increases speed as the task progresses. Since PACE depends on the probability distribution of the task's work requirement, we present methods for estimating this distribution and evaluate these methods on a variety of real workloads. We also show how to approximate the optimal schedule with one that changes speed a limited number of times. Using PACE causes very little additional overhead, and yields substantial reductions in CPU energy consumption. Simulations using real workloads show it reduces the CPU energy consumption of previously published algorithms by up to 49.5%, with an average of 20.6%, without any effect on performance.

Mismanagement of the persistent state of a system—all the executable files, configuration settings and other data that govern how a system functions—causes reliability problems, security vulnerabilities, and drives up operation costs. Recent research traces persistent state interactions—how state is read, modified, etc.—to help troubleshooting, change management and malware mitigation, but has been limited by the difficulty of collecting, storing, and analyzing the 10s to 100s of millions of daily events that occur on a single machine, much less the 1000s or more machines in many computing environments. We present the Flight Data Recorder (FDR) that enables always-on tracing, storage and analysis of persistent state interactions. FDR uses a domain-specific log format, tailored to observed file system workloads and common systems management queries. Our lossless log format compresses logs to only 0.5-0.9 bytes per interaction. In this log format, 1000 machine-days of logs—over 25 billion events—can be analyzed in less than 30 minutes. We report on our deployment of FDR to 207 production machines at MSN, and show that a single centralized collection machine can potentially scale to collecting and analyzing the complete records of persistent state interactions from 4000+ machines. Furthermore, our tracing technology is shipping as part of the Windows Vista OS. 1.

† HP Labs As the scope of personal data grows, it becomes increasingly difficult to find what we need when we need it. Desktop search tools provide a potential answer, but most existing tools are incomplete solutions: they index content, but fail to capture dynamic relationships from the user’s context. One emerging solution to this is contextenhanced search, a technique that reorders and extends the results of content-only search using contextual information. Within this framework, we propose using strict causality, rather than temporal locality, the current state of the art, to direct contextual searches. Causality more accurately identifies data flow between files, reducing the false-positives created by context-switching and background noise. Further, unlike previous work, we conduct an online user study with a fully-functioning implementation to evaluate user-perceived search quality directly. Search results generated by our causality mechanism are rated a statistically-significant 17 % higher on average over all queries than by using content-only search or context-enhanced search with temporal locality. 1

Increasingly, mobile computers use dynamic voltage scaling (DVS) to reduce CPU voltage and speed and thereby increase battery life. To determine how to change voltage and speed when responding to user interface events, we analyze traces of real user workloads. We evaluate a new heuristic for inferring when user interface tasks complete and find it is more efficient and nearly as effective as other approaches. We compare DVS algorithms and find that for a given performance level, the PACE algorithm uses the least energy and the Stepped algorithm uses the second least. We find that different types of user interface event (mouse movements, mouse clicks, and keystrokes) trigger tasks with significantly different CPU use, suggesting one should use different speeds for different event types. We also find differences in CPU use between categories of the same event type, e.g., between pressing spacebar and pressing enter, and between events of different applications. Thus, it is better to predict task CPU use based solely on tasks of the same category and application. However, energy savings from such improved predictions are small.

Personal data is growing at ever increasing rates, fueled by a growing market for personal computing solutions and dramatic growth of available storage space on these platforms. Users, no longer limited in what they can store, are now faced with the problem of organizing their data such that they can find it again later. Unfortunately, as data sets grow the complexity of organizing these sets also grows. This problem has driven a sudden growth in search tools aimed at the personal computing space, designed to assist users in locating data within their disorganized file space.

Machine learning techniques are widely used in many fields. One of the applications of machine learning in the field of the information security is classification of a computer behavior into malicious and benign. Anti viruses consisting on signature-based methods are helpless against new (unknown) computer worms. This paper focuses on the feasibility of accurately detecting unknown worm activity in individual computers while minimizing the required set of features collected from the monitored computer. A comprehensive experiment for testing the feasibility of detecting unknown computer worms, employing several computer configurations, background applications, and user activity, was performed. During the experiments 323 computer features were monitored by an agent that was developed. Four feature selection methods were used to reduce the amount of features and four learning algorithms were applied on the resulting feature subsets. The evaluation results suggests that using classification algorithms applied on only 20 features the mean detection accuracy exceeded 90%, and for specific unknown worms accuracy reached above 99%, while maintaining a low level of false positive rate.

In this paper we describe a Palm system call logging tool called Palmist. Palmist allows the practitioner to selectively collect statistics such as the system call invoked, application that invoked the system call, the time of the call and the call arguments. The logging mechanism adds a latency of about 10 msec per call to collect the log. On an average, the system uses about 20 bytes of memory on the PDA to store the log record. The mechanism has limitations in collecting logs for system calls that are needed by the collection mechanism itself. Our logging mechanism works for about 80% (707 of 881) of Palm OS 3.5 system calls. Our system can be utilized by system developers to customize their application behavior to optimize system parameters such as energy consumption, ease of use etc. 1.

In this paper we describe a Palm system call logging tool called Palmist. Palmist allows the practitioner to selectively collect statistics such as the system call invoked, application that invoked the system call, the time of the call and the call arguments. The logging mechanism adds a latency of about 10 msec per call to collect the log. On an average, the system uses about 20 bytes of memory on the PDA to store the log record. The mechanism has limitations in collecting logs for system calls that are needed by the collection mechanism itself. Our logging mechanism works for about 80% (707 of 881) of Palm OS 3.5 system calls. Our system can be utilized by system developers to customize their application behavior to optimize system parameters such as energy consumption, ease of use etc.

Wireless transmission of a bit can require over 1000 times more energy than a single 32-bit computation. It would therefore seem desirable to perform significant computation to reduce the number of bits transmitted. If the energy required to compress data is less than the energy required to send it, there is a net energy savings and consequently, a longer battery life for portable computers. This paper reports on the energy of lossless data compressors as measured on a StrongARM SA-110 system. We show that with several typical compression tools, there is a net energy increase when compression is applied before transmission. Reasons for this increase are explained, and hardwareaware programming optimizations are demonstrated. When applied to Unix compress, these optimizations improve energy efficiency by 51%. We also explore the fact that, for many usage models, compression and decompression need not be performed by the same algorithm. By choosing the lowest-energy compressor and decompressor on the test platform, rather than using default levels of compression, overall energy to send compressible web data can be reduced 31%. Energy to send harder-to-compress English text can be reduced 57%. Compared with a system using a single optimized application for both compression and decompression, the asymmetric scheme saves 11% or 12% of the total energy depending on the dataset.

This paper addresses algorithms for dynamically scaling CPU speed and voltage in order to save energy. Such scaling is useful and effective when the user will perceive the same performance, despite a slower CPU, as long as the task completes by some (soft) deadline. We show that it is possible to modify any scaling algorithm to minimize energy use without affecting perceived performance, and present a formula to do so. Because this formula specifies increased speed as the task progresses, we call this approach PACE (Processor Acceleration to Conserve Energy). This optimal formula depends on the probability distribution of the task's work requirement and requires that the speed be varied continuously.