Performance analysis of a distributed question/answering system

Surdeanu, Mihai; Moldovan, Dan; Harabagiu, Sanda M.

doi:10.1109/tpds.2002.1011413

Cited by 34 publications

(12 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This design follows previous proposals for complex search engines [24], where each node is able to compute queries autonomously. We illustrate an adaptation of the previously described architecture to a question answering (QA) system with three computing blocks (question processing, passage retrieval, and answer extraction) and its corresponding local caching pools in Figure 1.…”

Section: Related Workmentioning

confidence: 99%

Using Evolutive Summary Counters for Efficient Cooperative Caching in Search Engines

Domínguez-Sal

Aguilar-Saborit

Surdeanu

et al. 2012

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Abstract-We propose and analyze a distributed cooperative caching strategy based on the Evolutive Summary Counters (ESC), a new data structure that stores an approximated record of the data accesses in each computing node of a search engine. The ESC capture the frequency of accesses to the elements of a data collection, and the evolution of the access patterns for each node in a network of computers. The ESC can be efficiently summarized into what we call ESC-summaries to obtain approximate statistics of the document entries accessed by each computing node.We use the ESC-summaries to introduce two algorithms that manage our distributed caching strategy, one for the distribution of the cache contents, ESC-placement, and another one for the search of documents in the distributed cache, ESC-search. While the former improves the hit rate of the system and keeps a large ratio of data accesses local, the latter reduces the network traffic by restricting the number of nodes queried to find a document. We show that our cooperative caching approach outperforms state of the art models in both hit rate, throughput, and location recall for multiple scenarios, i.e., different query distributions and systems with varying degrees of complexity.

show abstract

Section: Related Workmentioning

confidence: 99%

Using Evolutive Summary Counters for Efficient Cooperative Caching in Search Engines

Domínguez-Sal

Aguilar-Saborit

Surdeanu

et al. 2012

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

show abstract

“…I/O operations in real systems can be either synchronous or asynchronous. It is assumed in our mechanism that all I/O operations are synchronous, because many I/O-intensive parallel applications issue synchronous read/write operations [Surdeanu et al 2002;Uysal et al 1997]. This assumption is conservative in the sense that it underestimates load balancing benefits (i.e., this assumption causes a number of undesired migrations with negative impact).…”

Section: Predicting Response Timementioning

confidence: 99%

Dynamic load balancing for I/O-intensive applications on clusters

et al. 2009

View full text Add to dashboard Cite

Load balancing for clusters has been investigated extensively, mainly focusing on the effective usage of global CPU and memory resources. However, previous CPU-or memory-centric load balancing schemes suffer significant performance drop under I/O-intensive workloads due to the imbalance of I/O load. To solve this problem, we propose two simple yet effective I/O-aware load-balancing schemes for two types of clusters: (1) homogeneous clusters where nodes are identical and (2) heterogeneous clusters, which are comprised of a variety of nodes with different performance characteristics in computing power, memory capacity, and disk speed. In addition to assigning I/O-intensive sequential and parallel jobs to nodes with light I/O loads, the proposed schemes judiciously take into account both CPU and memory load sharing in the system. Therefore, our schemes are able to maintain high performance for a wide spectrum of workloads. We develop analytic models to study mean slowdowns, task arrival, and transfer processes in system levels. Using a set of real I/O-intensive parallel applications and synthetic parallel jobs with various I/O characteristics, we show that our proposed schemes consistently improve the performance over existing non-I/O-aware load-balancing schemes, including CPU-and Memory-aware schemes and a PBS-like batch scheduler for parallel and sequential jobs, for a diverse set of workload conditions. Importantly, this performance improvement becomes much more pronounced when the applications are I/O-intensive. For example, the proposed approaches deliver 23.6-88.0 % performance improvements for I/Ointensive applications such as LU decomposition, Sparse Cholesky, Titan, Parallel text searching, and Data Mining. When I/O load is low or well balanced, the proposed schemes are capable of maintaining the same level of performance as the existing non-I/O-aware schemes.

show abstract

“…The algorithm stores a history record of the CPU and I/O time from previous queries and applies this record to estimate the fraction of CPU and I/O of the next computing block of the current query. The cost to compute the query q in node i is a weighted sum of the node load (Load and W CP U (q) ) [12]. We call this combined cost…”

Section: Load Balancingmentioning

confidence: 99%