76 research outputs found
Αποτελεσματική αναζήτηση πληροφορίας στον Παγκόσμιο Ιστό
No full textThe World Wide Web grows at alarming rates, making information retrieval an increasingly dificult process. Traditional search methods based on search-engines usually flood the users with an overwhelming number of URLs. If a user wants to stay up-to-date on some issue and he repeatedly queries the above search engines, s(h)e will be repeatedly flooded with (almost) the same set of URLs, out of which only a small percentage will point to new,previously unseen documents. In the first the first part of this thesis, a resource discovery tool built on top of traditional search engines and called USEwebNET. USEwebNET registers each user's interests and repeatedly queries several search engines for URLs matching a user's registered interests. USEwebNET keeps track of which URLs have been visited by each user. Thus, when a user invokes USEwebNET, he is presented only with new or "unvisited" URLs. The secord part of this thesis,focus in an extension of USEwebNET implemented especially for the area of digital libraries and called PaperFinder. PaperFinder, aims in helping scientists to keep track of the articles, which they are interested in among the growing number of papers, that become available on-line. It operates on top of popular Digital Libraries of scientific publications, filters only relevant papers delivers them to the users. PaperFinder may operate in a Keyword-based Mode, where scientists present a list of keywords that describe their field of interest, and in a Resource-discovery Mode, where scintists present one or more "seed papers", that describe a field of interest. In the latter, PaperFinder searches for papers that are revelant to htre seed paper and sorts the result by calculating their revelance to the seed paper.Ο Παγκόσμιος Ιστός μεγαλώνει με εκρηκτικούς ρυθμούς, πράγμα που κάνει την ανάκληση πληροφορίας μια ιδιαίτερη δύσκολη διαδικασία. Οι παραδοσιακές μέθοδοι αναζήτησης που στηρίζονται στις μηχανές αναζήτησης καταιγίζουν τους χρήστες με ένα υπερβολικά αριθμό URLs. Εάν ένας χρήστης επιθυμεί να παραμένει διαρκώς ενημερωμένος πάνω σε ένα θέμα, είναι αναγκασμένος να εκτελεί διαρκως τις ίδιες επερωτήσεις στις ίδιες μηχανές αναζήτησης,οι οποίες απαντούν με ένα σχεδόν όμοιο σύνολο από URLs. Από το σύνολο αυτό ένα μικρό ποσοστό δείχνει σε νέα πληροφορία. Στο πρώτο μέρος της εργασίας αυτής παρουσιάζεται ένα εργαλείο, που στηρίζεται πάνω σε ήδη υπάρχουσες μηχανές αναζήτησης και ονομάζεται USEwebNET. Το USEwebNET αποθηκεύει τα ενδιαφέροντα κάθε χρήστη και εκτελεί περιοδικά τις επερωτήσεις του στις υποστηριζόμενες μηχανές αναζήτησης. Παράλληλα, παρακολουθεί τα URLs, που επισκέπτεται ο χρήστης, και σε κάθε χρήση παρουσιάζει μόνο τα νέα ή μη αναγνωσμένα URLs. Το δεύτερο μέρος επικεντρώνεται σε μια επέκταση του USEwebNET, για το χώρο των Ψηφιακών Βιβλιοθηκών, που ονομάζεται PaperFinder. Το PaperFinder έχει σαν στόχο να βοηθήσει τον ερευνητή στην παρακολούθηση της πληροφορίας που τον ενδιαφέρει μέσα από το διαρκώς αυξανόμενο αριθμό των ερευνητικών άρθρων, που διατίθονται ηλεκτρονικά. Για το σκοπό αυτό, λειτουργεί πάνω από δημοφιλείς Ψηφιακές Βιβλιοθήκες, διαχωρίζει τα άρθρα και τα παραδίδει στους χρήστες. Υποστηρίζει δύο τρόπους αναζήτησης άρθρων. Ο πρώτος υποστηρίζει την εκτέλεση επερωτήσεων με τη χρήση λέξεων - κλειδιών - Keyword Base Mode - και ο δεύτερος στοχεύει στη γενίκευση των επερωτήσεων και αναταξινόμηση ως προς ένα ή περισσότερα "χαρακτηριστικά άρθρα", ώστε να βρεθεί μεγαλύτερο πλήθος δημοσιεύσεων - Resource Discovery Mode
Aggressive prefetching: an idea whose time has come
No full textAbstract I/O prefetching serves to hide the latency of slow pe-ripheral devices. Traditional OS-level prefetching strategies have tended to be conservative, fetching only thosedata that are very likely to be needed according to some simple heuristic, and only just in time for them to ar-rive before the first access. More aggressive policies, which might speculate more about which data to fetch, or fetch them earlier in time, have typically not beenconsidered a prudent use of computational, memory, or bandwidth resources. We argue, however, that techno-logical trends and emerging system design goals have dramatically reduced the potential costs and dramati-cally increased the potential benefits of highly aggressive prefetching policies. We propose that memory management be redesigned to embrace such policies. 1 Introduction Prefetching, also known as prepaging or read-ahead, hasbeen standard practice in operating systems for more than thirty years. It complements traditional cachingpolicies, such as LRU, by hiding or reducing the latency of access to non-cached data. Its goal is to predict futuredata accesses and make data available in memory before they are requested. A common debate about prefetching concerns how ag-gressive it should be. Prefetching aggressiveness ma
Knits: Switch-based connection hand-off
No full textAbstract—This paper describes a mechanism allowing nodes to hand-off active connections by utilizing connection splicing at an edge-switch serving as a gateway to a server cluster. The mechanism is primarily intended to be used as part of a content aware request distribution strategy. Our approach uses an extended form of network address translation which maps inbound connection information (ie. address, port, and sequence number) to a separate outbound connection. A key difference in our approach is that while the switch performs network address translation and TCP splicing, the actual hand-off is triggered by the back-end nodes. This relieves the switch of performing any application layer responsibilities. Nodes may hand-off connections by first initiating a new connection to the destination and then sending a message to the gateway which splices the two connections together. The gateway modifies subsequent packet headers in order to create a transparent hand-off. This mechanism requires no modification to the operating system on the servers or the clients and supports HTTP/1.1 persistent connections and pipelined requests. To test our design, we implemented a soft-switch using Linux Netfilter which includes the extended network address translation. We provide some preliminary performance analysis and make recommendations for future work. I
Lightweight Transactions on Networks of Workstations
No full textAlthough transactions have been a valuable abstraction of atomicity, persistency, and recoverability, they have not been widely used in programming environments today, mostly because of their high overheads that have been driven by the low performance of magnetic disks. A major challenge in transaction-based systems is to remove the magnetic disk from the critical path of transaction management. In this paper we present PERSEAS , a transaction library for main memory databases that decouples the performance of transactions from the magnetic disk speed. Our system is based on a layer of reliable main memory that provides fast and recoverable storage of data. We have implemented our system as a user-level library on top of the Windows NT operating system in a network of workstations connected with the SCI interconnection network. Our experimental results suggest that PERSEAS achieves performance that is orders of magnitude better than traditional recoverable main memory systems. 1 Intro..
Power-efficient Server-class Performance from Arrays of Laptop Disks
No full textThe disk array of a server-class system can account for a significant portion of the server’s total power budget. Similar observations for mobile (e.g. laptop) systems have led to the development of power management policies that spin down the hard disk when it is idle, but these policies do not transfer well to server-class disks. On the other hand, state-of-the-art laptop disks have response times and bandwidths within a factor of 2.5 of their server class cousins, and consume less than one sixth the energy. These ratios suggest the possibility of replacing a server-class disk array with a larger array of mirrored laptop disks. By spinning up a subset of the disks proportional to the current workload, we can exploit the latency tolerance and parallelism of typical server workloads to achieve significant energy savings, with equal or better peak bandwidth. Potential savings range from 50% to 80% of the total disk energy consumption
Abstract Competitive Prefetching for Data-Intensive Online Servers £
No full textRecent studies on operating system support for highly concurrent online servers mostly target CPU-intensive workloads with light disk I/O activities. However, an important class of online applications that access a large amount of disk-resident data, such as the index searching of large-scale Web search engines, has received limited attention. This paper examines operating system techniques to improve the throughput of data-intensive online servers under concurrent execution. Our contributions are two-fold. First, we propose a competitive prefetching strategy that balances the overhead of disk I/O switching and the wasted I/O bandwidth of prefetching unnecessary data. Second, we enhance the operating system memory management to reduce prefetching-incurred page thrashing at high concurrency levels. Our enhancements include the introduction of a new cache dedicated to prefetched pages and a novel page reclamation algorithm for it. We have implemented the proposed techniques in the Linux 2.6.3 kernel and conducted experiments based on microbenchmarks and three real applications (an index searching server, the Apache Web server, and a genetic sequence database). Our evaluation results demonstrate that competitive prefetching can improve the throughput of real applications by up to 47 % at low concurrency while the enhanced memory management scheme can produce up to a four-fold performance enhancement at high concurrency. The performance improvement is achieved without any application changes.
Increasing Disk Burstiness for Energy Efficiency
No full textHard disks for portable devices, and the operating systems that manage them, incorporate spin-down policies that idle the disk after a certain period of inactivity. In essence, these policies use a recent period of inactivity to predict that the disk will remain inactive in the near future. We propose an alternative strategy, in which the operating system deliberately seeks to cluster disk operations in time, to maximize the utilization of the disk when it is spun up and the time that the disk can be spun down. In order to cluster disk operations we postpone the service of non-urgent operations, and use aggressive prefetching and file prediction to reduce the likelihood that synchronous reads will have to go to disk. In addition, we present a novel predictive spin-down/spin-up policy that exploits high level operating system knowledge to decrease disk idle time prior to spin-down, and application wait time due to spin-up. We evaluate our strategy through trace-driven simulation of several different workload scenarios. Our results indicate that the deliberate creation of bursty activity can save up to 55% of the energy consumed by an IBM TravelStar disk, while simultaneously decreasing significantly the negative impact of disk spin-up latency on application performance
KNITS: Switch-based Connection Hand-off
No full textThis paper describes a mechanism allowing nodes to hand-off active connections by utilizing connection splicing at an edge-switch serving as a gateway to a server cluster. The mechanism is primarily intended to be used as part of a content aware request distribution strategy. Our approach uses an extended form of network address translation which maps inbound connection information (ie. address, port, and sequence number) to a separate outbound connection. A key difference in our approach is that while the switch performs network address translation and TCP splicing, the actual hand-off is triggered by the back-end nodes. This relieves the switch of performing any application layer responsibilities. Nodes may hand-off connections by first initiating a new connection to the destination and then sending a message to the gateway which splices the two connections together. The gateway modifies subsequent packet headers in order to create a transparent hand-off. This mechanism requires no modification to the operating system on the servers or the clients and supports HTTP/1.1 persistent connections and pipelined requests. To test our design, we implemented a soft-switch using Linux Netfilter which includes the extended network address translation. We provide some preliminary performance analysis and make recommendations for future work
- …
