Concise Papers. Main Memory Indexing: The Case for BD-Tree 1 INTRODUCTION 3 COST ANALYSIS 2 THE MEMORY-BASED BD-TREE

Size: px
Start display at page:

Download "Concise Papers. Main Memory Indexing: The Case for BD-Tree 1 INTRODUCTION 3 COST ANALYSIS 2 THE MEMORY-BASED BD-TREE"

Transcription

1 870 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 16, NO. 7, JULY 2004 Concise Papers Main Meory Indexing: The Case for BD-Tree BinCui,BengChinOoi,Meber, IEEE, Jianwen Su, Senior Meber, IEEE, and Kian-Lee Tan, Meber, IEEE Coputer Society Abstract In this paper, we adapt and optiize the BD-tree for ain eory data processing. We copare the eory-based BD-tree against the B þ -tree and CSB þ -tree. We present cost odes for exact atch uery for these indexes, incuding L2 cache and transation ookahead buffer (TLB) iss ode and execution tie ode. We aso ipeented these structures and conducted experienta study. Our anaytica and experienta resuts show that a we-tuned BD-tree is superior in ost cases. Index Ters Main eory databases, BD-tree, B þ -tree, CSB þ -tree. 1 INTRODUCTION æ THE ipact of advances in hardware and ain eory capacity is beginning to be fet by the software counity [3], [4], [6], [7], [12]. Not ony is it possibe to store the entirety of a database in eory now, any agoriths that were designed in the past need to be reexained. For exape, the eory-based index structure T -tree [10] is no onger attractive because of its ow fanout, poor cache behavior, and excessive utiization of pointers. In this paper, we reexaine the indexing issue in ain eory databases. To iprove perforance, an indexing structure ust faciitate effective use of the L2 caches and CPU. One proising structure is the CSB þ -tree [12] that reduces the cache isses by storing a the chid nodes of any given node contiguousy, and keeping ony the address of the first chid in each node, thereby increasing the utiization of the cache ine. However, whie the CSB þ -tree is efficient for range ueries, it does not perfor we for exact atch ueries (copared to hash-based schees). This is because utipe nodes need to be accessed during the tree traversa. This propted us to search for a structure that can cobine the benefits of hashing and tree indexing. In this paper, we present our soution to faciitate efficient uery processing in ain eory environent. We adapted and optiized the Bounded Disorder (BD) ethod introduced by Litwin and Loet [11] for ain eory processing. The BD-tree is essentiay a B þ -tree where the eaf nodes are arge-sized partitions, and each eaf node is organized by hash tabes. We evauated the optiized eory-based BD-tree against the CSB þ -tree and B þ -tree anayticay and epiricay. Our resuts show that a we-tuned BD-tree outperfors the other two indexes in ost cases. 2 THE MEMORY-BASED BD-TREE Fig. 1 shows the structure of the BD-tree [11], which coprises two tiers. In the first tier, a hierarchica utiway tree structure (e.g.,. B. Cui, B.C. Ooi, and K.-L. Tan are with the Departent of Coputer Science, Nationa University of Singapore, 3 Science Drive 2, Singapore E-ai: {cuibin, ooibc, tank}@cop.nus.edu.sg.. J. Su is with the Departent of Coputer Science, University of Caifornia at Santa Barbara, Santa Barbara, CA E-ai: su@cs.ucsb.edu. Manuscript received 1 Oct. 2002; revised 11 May 2003; accepted 28 Juy For inforation on obtaining reprints of this artice, pease send e-ai to: tkde@coputer.org, and reference IEEECS Log Nuber /04/$20.00 ß 2004 IEEE Pubished by the IEEE Coputer Society B þ -tree) range partitions the data. The second tier begins at the eaf eve of the tree, which organizes the data in each eaf node using a hashing ethod. Records within a eaf are not ordered. The BD-tree is efficient for eory-based uery processing. First, the eaf node size of a BD-tree is uch arger than that of a B þ -tree. This eans fewer interna nodes are needed, i.e., a shorter tree. Therefore, fewer cache and TLB 1 isses are needed during tree traversa. Moreover, since eaf nodes are organized as hash buckets, ony the necessary buckets need to be searched. This ceary benefits exact atch ueries. Second, the tree structure essentiay acts as a dynaic range partitioning echanis, aowing each eaf node to have different range size. As such, ony the necessary eaves need to be exained for range ueries. To optiize the BD-tree for ain eory processing, we adapt the hash function with ow coputation cost to reduce the CPU overhead of searching the eaf nodes. Since the use of tree structure essentiay acts as a dynaic range partitioning echanis and the difference between the nubers of keys in the eaf nodes is bounded, the average perforance can be accuratey predicted under the unifority assuption. Moreover, since data ay be skewed or the hash functions ay not distribute the data unifory within each eaf, we aow each bucket to have at ost k chains (for soe integer k>¼ 1). In other words, a eaf wi be spit ony when a new record is to be inserted into a bucket whose k chained buckets are aready fu. There are exceptions where the dupicated keys (hash keys) overfow the chained buckets and no spitting wi hep. In these cases, we aow ore than k chains. For sipicity, we ignore such specia cases in our discussion. Since the fanout of a ain eory tree structure is typicay sa copared to a diskbased index, we can reduce the height of BD-tree significanty by setting a reativey arge eaf size. Within the eaf, we set the bucket size to correspond to the cache ine size. Note that, athough we can keep the eaf nodes of a B þ -tree=csb þ -tree arger than its interna nodes, a arger eaf node size does not resut in better search perforance. Because a arge eaf node size aso eans ore cache isses when the node is accessed, soe noncontiguous segents of the eaf need to be oaded into cache, even if we use binary search. For exact atch ueries in the BD-tree, we ony need one cache iss to access each chain bucket and at ost k buckets to be retrieved in eaf eve. For range ueries, because the eaf partitions are ordered, we just scan the eaves that overap the desired uery ranges. To speed up the range uery, we store the chained buckets contiguous with the priary buckets of the eaf in the eory and, hence, reduce the TLB isses of scanning. 3 COST ANALYSIS To have an insight into the perforance of the B þ -tree, CSB þ -tree, and BD-tree, we present anaytica odes to evauate the cost of exact atch uery, i.e., a cache and TLB isses ode and an execution tie ode. We ode the cost of exact atch uery as a function of three variabes: nuber of L2 cache isses, TLB isses, and instructions. There are other factors that affect perforance, such as branch ispredictions, instruction cache isses, and L1 data cache isses. However, these factors do not pay a significant roe in deterining the overa cost [3]. The representative paraeters used for the anaysis are shown in Tabe 1. Soe of the vaues were produced by easuring the perforance of indexes in the experient. And, to sipify the anaysis, we assue that the buckets in the eaf nodes of the BD-tree have no overfow chains. 1. The TLB is an essentia part of odern processors, which keeps the apping fro ogica eory address to physica eory address.

2 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 16, NO. 7, JULY Fig. 1. The BD-tree. 3.1 Cache and TLB Miss Mode The L2 cache isses and TLB isses pay an iportant roe in the tree operations. For each node access, there ay be severa cache isses and one TLB iss for various node sizes (< virtua eory page size). The node size and tree height are two ain paraeters in this cost ode. A arger node size ipies a shorter tree, and vice versa. To deterine the height of the tree, we first present the forua for fanout, f. 8 j k < u jnj jj þ 1 : CSB þ -tree f ¼ j k : u jnj jj þjpj þ 1 : B þ -tree=bd-tree: Aso, with the given size of data set, the nuber of eaf node, L, is: 8 c >< : B þ -tree=csb þ -tree u jnj jj L ¼ >: c ud jbj : BD-tree: Therefore, the height of the tree, h, is given as foows: h ¼dog f L þ 1e: For an exact atch uery, we ust traverse the tree fro the root to the eaf to get the answer. At each eve, a chid node is deterined by a binary search in the parent node, so the nuber of L2 cache ines accessed of a node is dðog 2 ðu n c Þþ0:5Þe, where n c is the nuber of cache ines spanned by one node (n c ¼d jnj jj e). Because we aways need to access the first cache bock of the node, we add 0.5 cache iss for each node access. On the first uery of the index, the nuber of cache isses is the cache ines accessed. However, there is a higher probabiity of finding soe nodes in the cache for subseuent ueries, e.g., the nodes in the upper eves of the tree are accessed freuenty and ay aways reside in the cache. Reaccessing of these nodes wi resut in fewer cache isses. To ode cache isses ore efficienty, we use the Cardenas s forua [5]: Xðn; Þ ¼n ð1 ð1 1=nÞ Þ; where X is the nuber of uniue nodes accessed, n is the nuber of nodes avaiabe, and is the nuber of ueries. Using this forua, the average nuber of uniuey accessed nodes for a uery, i.e. Xðn;Þ, reduces when we increase the nuber of ueries. We cobine the Cardenas s forua with the tree structure to deterine the nuber of uniue nodes that are accessed when we traverse the tree. The nuber of cache isses in the eaf node of BD-tree is 1, as we assue no chained buckets in the ode. Thus, we can ode the nuber of cache isses per uery as: TABLE 1 Paraeters for Anaysis

3 872 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 16, NO. 7, JULY 2004 Fig. 2. The nuber of (a) cache isses and (b) TLB isses for a singe uery. 8 P h >< XðnðiÞ;Þdðog i¼1 2ðun cþþ0:5þe : B þ -tree=csb þ -tree C ¼ P h 1 >: XðnðiÞ;Þdðog i¼1 2ðun cþþ0:5þeþxðnðhþ; Þ : BD-tree; where nðiþ is the nuber of nodes at eve i in the tree (nðiþ ¼f i 1 ). The ode gives the copusory cache isses, but not the capacity cache isses [9]. Because the cache size is finite, eventuay, soe highy accessed cache ines can be repaced after running any ueries. However, with arge cache size, these effects are not significant for a first-order approxiation. For an individua uery, the nuber of TLB isses can be eua to the height of the tree, whie the TLB isses ay be satisfied fro the L2 cache hits and top eves of the tree ay stay cached for subseuent ueries. Therefore, the nuber of the TLB isses, TLB, can be uch saer than the tree height and aso saer than C because of arger TLB page size. TLB is represented by the foowing euation: TLB ¼ ( C og 2 ðun cþþ0:5 : B þ -tree=csb þ -tree C 1 og 2 ðun þ 1 cþþ0:5 : BD-tree: We show the average nuber of TLB and cache isses for 1,000 exact atch ueries using the above odes in Fig. 2. The B þ -tree perfors worst aong these trees. Copared with the CSB þ -tree, the B þ -tree has a saer branching factor than the CSB þ -tree since it needs to store ore chid pointers. Hence, the B þ -tree is taer than the CSB þ -tree, and incurs ore TLB and cache isses. Not surprisingy, the BD-tree is best because its eaf node (partition) has ore keys, thus the BD-tree is shorter. When we increase the node size, both the nuber of cache and TLB isses reduce. The nuber of cache isses is uch saer than the nuber of cache ines accessed, because soe of the freuenty accessed nodes can reside in the cache. The nuber of TLB isses is fewer than that of cache isses when the node size is arger than 64 bytes, because one node access ony incurs one TLB iss, but severa cache isses in this case. For the CSB þ -tree, when the node size is 64 bytes (cache ine size), the nuber of cache isses is saest. For the B þ -tree and BD-tree, the optia node size is 128 bytes. In a cases, the cache isses do not increase uch for arger node sizes. Because the tree is shorter, ore proportiona accessed nodes can be resident in the cache, athough ore cache ines need to be accessed. The nuber of cache isses is a coproise of these factors. 3.2 Execution Tie Mode The tie to execute a singe uery (T Q ) incudes coputation tie (T C ), L2 cache isses tie (T CM ), and TLB isses tie (T TLBM ). Thus, we can estiate the execution tie T Q as: T Q ¼ T C þ T CM þ T TLBM : The above euation ignores the effects of out-of-order execution, which can hide a portion of cache iss atency by continuing to execute the instructions out-of-order on the processor [3]. It is easy for us to get T CM and T TLBM because we know the nuber of TLB and cache isses fro the TLB and cache iss ode. We obtain the cost of each TLB and cache iss using the Caibrator Too [1], e.g., a L2 cache iss atency c is 150 cyces and a TLB iss atency t is 100 cyces for SUNFire The coputation cost is the tie taken to execute a the instructions for a uery. The nuber of instructions executed at each eve is ainy the binary search inside the node, so we can give the euation to cacuate the nuber of tota instructions I Q as foows: I Q ¼ 8 I b ðh 1Þ og 2 u jnj jj þjpj og 2 u jnj jj þ 1 >< I b ðh 1Þ og 2 u jnj jj og 2 u jnj jj þ 1 I b ðh 1Þ og 2 u jnj jj >: og 2 u jbj þ 1 : B þ -tree : CSB þ -tree : BD-tree; where I b is the nuber of instructions reuired to evauate a key and seect the next position in a binary search. The Perfon too [2] can be used to count the nuber of instructions/cyces of an evauation. Hence, we can get the nuber of instructions for an exact atch uery using the above euation. The CPU cyces per instruction (CPI) used in the ode is eua to 2. The nuber of instructions and tota CPU cyces for a uery are shown in Fig. 3. Fig. 3a shows the nuber of instructions executed versus node size per uery. The nuber of instructions is uite stabe when the node size is arger than 64 bytes because we execute binary search in each node and the overa coputation cost is siiar. More instructions are incurred when the node size is eua to 32 bytes because the binary search is not efficient when the node size is too

4 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 16, NO. 7, JULY Fig. 3. The nuber of instructions and cyces for a singe uery. (a) Instructions and (b) CPU cyces. sa. The tota execution tie is the cobination of TLB and cache isses and CPU cost. When the node size is eua to the cache ine size, the CSB þ -tree perfors best, whie 128 bytes is an optia node size for B þ -tree and BD-tree. Ceary, the BD-tree perfors best aong these trees, foowed by the CSB þ -tree and B þ -tree. 4 APERFORMANCE STUDY In this section, we present an experienta evauation on SUNFire 4800, which has 750MHz CPU, 16 GB RAM, and 8M L2 cache. The data set we used consists of 10 iion uniue integers. The data in the structures consists of < key; pointer > pairs, both 4 bytes ong, and the keys are unifory distributed. We aso conducted sensitivity anaysis on the techniues by varying distributions of data sets, cardinarity of data sets, and so on. For a fu set of of the experienta study, readers are referred to [8]. The BD-tree has two extra paraeters, the nuber of buckets in eaf and the bucket size. Our preiinary study to tune these paraeters shows that the exact atch uery perforance of BDtree iproves with a arge nuber of buckets. On the other hand, the arge eaf size ay reduce the perforance of range ueries with sa uery range. We found that the perforance of tree is near optia when the eaf has 64 buckets, each of size 64 bytes, and we use it as the defaut vaues in our study here. Moreover, we aow at ost a chain of two buckets (i.e., k ¼ 2). A the buckets in a eaf node are aocated in the contiguous eory space to iniize TLB isses. 4.1 On Exact Match Query In this experient, we study the effect of node size on the perforance of exact atch uery of the various schees by varying the node size fro 32 bytes to 256 bytes. This aso serves to tune the various schees for optia settings of node sizes. We conducted 1,000 operations of exact atch ueries, and recorded their perforance. The resuts are shown in Fig. 4, and it is cear that the perforance of the three indexes is consistent with the anaytica resuts. Fig. 4a shows the overa eapsed tie of exact atch ueries. For a tree-based schees, when the node size is too sa (e.g., 32 bytes), the fanout becoes so sa that it resuts in a very ta tree. This eads to ore TLB and cache isses as the tree is traversed. When the node size increases, the nuber of TLB and cache isses decreases. For the B þ -tree, its optia perforance for exact atch ueries occurs when the node size is eua to 128 bytes. For the CSB þ -tree, we observe that it is optia when the node size is 64 bytes which is the cache ine size of L2. Siiar to the B þ -tree, the BD-tree yieds optia perforance when the interna node size is eua to 128 bytes. Coparing these various B þ -tree-based structures, we see that the CSB þ -tree is better than the B þ -tree in a cases, but the difference is not significant when the node size is arger than 128 bytes. This resut is consistent with Fig. 4. Perforance of exact atch uery. (a) Eapsed tie of exact atch and (b) cache isses of exact atch.

5 874 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 16, NO. 7, JULY 2004 Fig. 5. Perforance of range uery and insertion. (a) Eapsed tie of range uery and (b) eapsed tie of insertion. the findings in the anaytica study. The BD-tree perfors best and is about 40 percent better because of the shorter tree traversa and the efficient bucket access in the eaf eve. In Fig. 4b, we show the nubers of average cache isses of exact atch uery, which were obtained by Perfon [2]. Because we did not fush the L2 cache after each uery and the L2 cache can be fairy arge on odern achines, running any ueries on the index wi eventuay resut in fewer cache isses per uery, e.g., soe highy accessed cache ines can aways reside in the L2 cache. Note that there is no index node residing in the cache before the operations. So, saer index size can benefit ore fro this echanis as the upper eves of the tree can aways reside in the cache. Based on the resuts of the previous experient, we pick the node size for the BD-tree and B þ -tree to be 128 bytes, the node size for the CSB þ -tree to be 64 bytes. These wi be used as the defaut settings for the subseuent experients. 4.2 On Range Query In this experient, we study the various schees perforance on range ueries. We vary the seectivities of the ueries ranges fro 0.01 percent to 10 percent of the tota nuber of keys. The resut is shown in Fig. 5a. The cost of range ueries fas into two parts, tree traversa to ocate the eaf node containing the first record and scanning a eaf nodes that contain the answers. The CSB þ -tree is aways better than the B þ -tree because the CSB þ -tree can ocate the eaf faster and fewer TLB isses are incurred for scan, as the eaf nodes with the sae parent node are stored seuentiay in eory. Athough the BD-tree can ocate the eaf node uch faster, the disorder within the eaf incurs soe additiona cost. Therefore, when the range seectivity is sa, e.g., 0.01 percent seectivity, the BD-tree shows poorer resuts because it needs to access the whoe eaf. On the other hand, the BD-tree perfors as we as the CSB þ -tree and better than the B þ -tree when the seectivity is arger than 0.1 percent in our experient. Each eaf node of the BD-tree has uch ore keys than the other two trees and, hence, the scan of eaf nodes causes fewest TLB isses. 4.3 On Insertion We aso study the insertion perforance of the various schees. Fig. 5b shows the resut of 1,000 insertion for various node sizes. In a cases, we observe siiar perforance: the BD-tree is the ost efficient, the CSB þ -tree is the worst and the B þ -tree is sighty better than the CSB þ -tree. The CSB þ -tree is worse than the B þ -tree for the foowing reasons. The insertion cost has three parts: search cost, sort cost, and the spit cost. The spit cost of the CSB þ -tree incudes copying a copete node group, whereas a node spit of the B þ -tree invoves creating a new node. The BD-tree perfors better than the B þ -tree because the eaf partitions of the BD-tree have uch ore keys than those of the CSB þ -tree and B þ -tree, thus the tree is shorter; at the sae tie, inserting a key within a eaf is aso very efficient, as we ony need to cacuate the hash vaue and insert the key into the target bucket. 5 CONCLUSION In this paper, we have revisited the probe of accessing data in ain eory databases. We adapted and optiized the BD-tree [11] to faciitate fast search in ain eory environent. The BD-tree taps on the strengths of a hierarchica tree structure to eveny distribute and range partition the keys (faciitating efficient range ueries processing) and the efficiency of hash-based ethods for processing exact atch ueries. We studied the BD-tree against the B þ -tree and CSB þ -tree anayticay and epiricay. Our resuts showed that the BD-tree is a proising index structure for eory-based processing. REFERENCES [1] The Caibrator Too, [2] The Perfon Too, [3] A. Aiaaki, D.J. DeWitt, M.D. Hi, and D.A. Wood, DBMS on a Modern Processor: Where Does Tie Go, Proc. 25th Very Large Databases Conf., pp , [4] P. Bohannon, P. Mcroy, and R. Rastogi, Main-Meory Index Structures with Fixed-Size Partia Keys, Proc. ACM SIGMOD, pp , [5] A. Cardenas, Anaysis and Perforance of Inverted Database Structures, Co. ACM, vo. 18, pp , May [6] S. Chen, P.B. Gibbons, and T.C. Mowry, Iproving Index Perforance through Prefetching, Proc. ACM SIGMOD, pp , [7] S. Chen, P.B. Gibbons, T.C. Mowry, and G. Vaentin, Fracta Prefeching B+-Tree: Optiizing Both Cache and Disk Perforance, Proc. ACM SIGMOD, pp , [8] B. Cui, B.C. Ooi, J.W. Su, and K.L. Tan, Using BD-Tree for Main Meory Processing, technica report, Schoo of Coputing, Nat Univ. of Singapore, [9] J. Hennessy and D. Patterson, Coputer Architecture: A Quantitative Approach. Morgan Kauffan, [10] T. Lehan and M. Carey, A Study of Index Structures for Main Meory Database Manageent Systes, Proc. 12th Very Large Databases Conf., pp , [11] W. Litwin and D. Loet, The Bounded Disorder Access Method, Proc. 17th Int Conf. Data Eng., pp , [12] J. Rao and K. Ross, Making B+-Trees Cache Conscious in Main Meory, Proc. ACM SIGMOD, pp , For ore inforation on this or any other coputing topic, pease visit our Digita Library at

Performance Modeling of Database Servers in a Telecommunication Service Management System

Performance Modeling of Database Servers in a Telecommunication Service Management System ICDT : The Seventh Internationa Conference on Digita Teecounications Perforance Modeing of Database Servers in a Teecounication Service Manageent Syste Maria Kih, Paya Aani, Anders Robertsson, Gabriea

More information

Database Replication Algorithm Performance in High Speed Networks Under Load Balancing

Database Replication Algorithm Performance in High Speed Networks Under Load Balancing Database Repication Agorith Perforance in High Speed Networks Under Load Baancing Rekh Nath Singh 1, Raghura Singh 2 1 Research Schoar, A. P. J. Abdu Kaa Technica University, Lucknow, India. 2 Director,

More information

A NEW METHOD FOR OPTIMAL LOCATION OF FACTS CONTROLLERS USING GENETIC ALGORITHM

A NEW METHOD FOR OPTIMAL LOCATION OF FACTS CONTROLLERS USING GENETIC ALGORITHM Journa of heoretica and Appied Inforation echnoogy 200-2007 JAI. A rights reserved. www.atit.org A NEW MEHOD FOR OPIMAL LOCAION OF FACS CONROLLERS USING GENEIC ALGORIHM 1 K. Vayakuar, 2 Dr. R. P. Kuudinidevi

More information

A Fast Recovery Technique for Multi-Point to Multi-Point MPLS tunnels

A Fast Recovery Technique for Multi-Point to Multi-Point MPLS tunnels M. Chaitou and J. L. Roux / IJECCT 01, Vo. (3) 34 A Fast Recovery Technique for Muti-Point to Muti-Point MPLS tunnes Mohaad Chaitou and Jean-Louis Le RouxOrange Labs avenue Pierre Marzin, 300 Lannion France

More information

Improving Memory Energy Using Access Pattern Classification

Improving Memory Energy Using Access Pattern Classification Iproving Meory Energy Using Access Pattern Cassification Mahut Kandeir Microsystes Design Lab Pennsyvania State University University Par, PA 16802 andeir@cse.psu.edu Ugur Sezer ECE Departent University

More information

Origami Axioms. O2 Given two marked points P and Q, we can fold a marked line that places P on top of Q.

Origami Axioms. O2 Given two marked points P and Q, we can fold a marked line that places P on top of Q. Origai Axios Given a piece of paper, it is possibe to fod ots of different ines on it. However, ony soe of those ines are constructibe ines, eaning that we can give precise rues for foding the without

More information

Running Tite: Conict-Free Access of Paths Address for Correspondence: M.C. Pinotti IEI-CNR Via S. Maria, Pisa ITALY E-ai:

Running Tite: Conict-Free Access of Paths Address for Correspondence: M.C. Pinotti IEI-CNR Via S. Maria, Pisa ITALY E-ai: Mappings for Conict-Free Access of Paths in Bidiensiona Arrays, Circuar Lists, and Copete Trees Aan A. Bertossi y and M. Cristina Pinotti Istituto di Eaborazione de' Inforazione Nationa Counci of Research

More information

DETERMINING INTUITIONISTIC FUZZY DEGREE OF OVERLAPPING OF COMPUTATION AND COMMUNICATION IN PARALLEL APPLICATIONS USING GENERALIZED NETS

DETERMINING INTUITIONISTIC FUZZY DEGREE OF OVERLAPPING OF COMPUTATION AND COMMUNICATION IN PARALLEL APPLICATIONS USING GENERALIZED NETS DETERMINING INTUITIONISTIC FUZZY DEGREE OF OVERLAPPING OF COMPUTATION AND COMMUNICATION IN PARALLEL APPLICATIONS USING GENERALIZED NETS Pave Tchesmedjiev, Peter Vassiev Centre for Biomedica Engineering,

More information

Performance Analysis of RAID in Different Workload

Performance Analysis of RAID in Different Workload Send Orders for Reprints to reprints@benthascience.ae 324 The Open Cybernetics & Systeics Journal, 2015, 9, 324-328 Perforance Analysis of RAID in Different Workload Open Access Zhang Dule *, Ji Xiaoyun,

More information

Extending Graph Rewriting for Refactoring

Extending Graph Rewriting for Refactoring Extending Graph Rewriting for Refactoring Nies Van Eetvede, Dirk Janssens University of Antwerp Departent of oputer science Middeheiaan 1 2020 Antwerpen {nies.vaneetvede dirk.janssens@ua.ac.be Abstract.

More information

A Secure Approach for Caching Contents in Wireless Ad Hoc Networks

A Secure Approach for Caching Contents in Wireless Ad Hoc Networks A Secure Approach for Caching Contents in Wireess Ad Hoc Networks Mohsen Karizadeh Kiskani and Haid R. Sadjadpour Abstract Caching ais to store data ocay in soe nodes within the network to be abe to retrieve

More information

MosaicShape: Stochastic Region Grouping with Shape Prior

MosaicShape: Stochastic Region Grouping with Shape Prior Boston University Coputer Science Technica Report No. 2005-008, Feb. 2005. To appear in Proc. CVPR, 2005. MosaicShape: Stochastic Region Grouping with Shape Prior Jingbin Wang Erdan Gu Margrit Bete Coputer

More information

Lines and Angles. introduction

Lines and Angles. introduction 9 Lines and nges intrductin In cass VI, you have earnt soe basic concepts and ters of geoetry point, ine, pane, ine segent, ray, ange and types of anges. In this chapter, we sha earn about soe pairs of

More information

PARALLEL database systems are essential to important

PARALLEL database systems are essential to important IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 13, NO. 12, DECEMBER 2002 1211 Load Baanced and Optia Disk Aocation Strategy for Partia Match Queries on Mutidiensiona Fies Saja K. Das, Meber,

More information

Grading Results Total 100

Grading Results Total 100 University of California, Berkeley College of Engineering Departent of Electrical Engineering and Coputer Sciences Fall 2003 Instructor: Dave Patterson 2003-11-19 v1.9 CS 152 Exa #2 Solutions Personal

More information

Collaborative Web Caching Based on Proxy Affinities

Collaborative Web Caching Based on Proxy Affinities Collaborative Web Caching Based on Proxy Affinities Jiong Yang T J Watson Research Center IBM jiyang@usibco Wei Wang T J Watson Research Center IBM ww1@usibco Richard Muntz Coputer Science Departent UCLA

More information

An Efficient Approach for Content Delivery in Overlay Networks

An Efficient Approach for Content Delivery in Overlay Networks An Efficient Approach for Content Delivery in Overlay Networks Mohaad Malli, Chadi Barakat, Walid Dabbous Projet Planète, INRIA-Sophia Antipolis, France E-ail:{alli, cbarakat, dabbous}@sophia.inria.fr

More information

Implementation of fast motion estimation algorithms and comparison with full search method in H.264

Implementation of fast motion estimation algorithms and comparison with full search method in H.264 IJCSNS International Journal of Coputer Science and Network Security, VOL.8 No.3, March 2008 139 Ipleentation of fast otion estiation algoriths and coparison with full search ethod in H.264 A.Ahadi, M.M.Azadfar

More information

A Memory Grouping Method for Sharing Memory BIST Logic

A Memory Grouping Method for Sharing Memory BIST Logic A Memory Grouping Method for Sharing Memory BIST Logic Masahide Miyazai, Tomoazu Yoneda, and Hideo Fuiwara Graduate Schoo of Information Science, Nara Institute of Science and Technoogy (NAIST), 8916-5

More information

Language Identification for Texts Written in Transliteration

Language Identification for Texts Written in Transliteration Language Identification for Texts Written in Transiteration Andrey Chepovskiy, Sergey Gusev, Margarita Kurbatova Higher Schoo of Economics, Data Anaysis and Artificia Inteigence Department, Pokrovskiy

More information

Detection of Outliers and Reduction of their Undesirable Effects for Improving the Accuracy of K-means Clustering Algorithm

Detection of Outliers and Reduction of their Undesirable Effects for Improving the Accuracy of K-means Clustering Algorithm Detection of Outliers and Reduction of their Undesirable Effects for Iproving the Accuracy of K-eans Clustering Algorith Bahan Askari Departent of Coputer Science and Research Branch, Islaic Azad University,

More information

Database Design on Mechanical Equipment Operation Management System Zheng Qiu1, Wu kaiyuan1, Wu Chengyan1, Liu Lei2

Database Design on Mechanical Equipment Operation Management System Zheng Qiu1, Wu kaiyuan1, Wu Chengyan1, Liu Lei2 2nd International Conference on Advances in Mechanical Engineering and Industrial Inforatics (AMEII 206) Database Design on Mechanical Equipent Manageent Syste Zheng Qiu, Wu kaiyuan, Wu Chengyan, Liu Lei2

More information

Investigation of The Time-Offset-Based QoS Support with Optical Burst Switching in WDM Networks

Investigation of The Time-Offset-Based QoS Support with Optical Burst Switching in WDM Networks Investigation of The Tie-Offset-Based QoS Support with Optical Burst Switching in WDM Networks Pingyi Fan, Chongxi Feng,Yichao Wang, Ning Ge State Key Laboratory on Microwave and Digital Counications,

More information

The optimization design of microphone array layout for wideband noise sources

The optimization design of microphone array layout for wideband noise sources PROCEEDINGS of the 22 nd International Congress on Acoustics Acoustic Array Systes: Paper ICA2016-903 The optiization design of icrophone array layout for wideband noise sources Pengxiao Teng (a), Jun

More information

PERFORMANCE MEASURES FOR INTERNET SERVER BY USING M/M/m QUEUEING MODEL

PERFORMANCE MEASURES FOR INTERNET SERVER BY USING M/M/m QUEUEING MODEL IJRET: International Journal of Research in Engineering and Technology ISSN: 239-63 PERFORMANCE MEASURES FOR INTERNET SERVER BY USING M/M/ QUEUEING MODEL Raghunath Y. T. N. V, A. S. Sravani 2 Assistant

More information

COLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL

COLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL COLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL 1 Te-Wei Chiang ( 蔣德威 ), 2 Tienwei Tsai ( 蔡殿偉 ), 3 Jeng-Ping Lin ( 林正平 ) 1 Dept. of Accounting Inforation Systes, Chilee Institute

More information

Gromov-Hausdorff Distance Between Metric Graphs

Gromov-Hausdorff Distance Between Metric Graphs Groov-Hausdorff Distance Between Metric Graphs Jiwon Choi St Mark s School January, 019 Abstract In this paper we study the Groov-Hausdorff distance between two etric graphs We copute the precise value

More information

l Tree: set of nodes and directed edges l Parent: source node of directed edge l Child: terminal node of directed edge

l Tree: set of nodes and directed edges l Parent: source node of directed edge l Child: terminal node of directed edge Trees & Heaps Week 12 Gaddis: 20 Weiss: 21.1-3 CS 5301 Fa 2016 Ji Seaman 1 Tree: non-recursive definition Tree: set of nodes and directed edges - root: one node is distinguished as the root - Every node

More information

Game Theory-Based Nonlinear Bandwidth Pricing for Congestion Control in Cloud Networks

Game Theory-Based Nonlinear Bandwidth Pricing for Congestion Control in Cloud Networks 216 IEEE 8th Internationa Conference on Coud Coputing Technoogy and Science Gae Theory-ased Noninear andwidth Pricing for Congestion Contro in Coud Networs Abouzar Ghavai, Zhuozhao Li and Haiying Shen

More information

Special Edition Using Microsoft Excel Selecting and Naming Cells and Ranges

Special Edition Using Microsoft Excel Selecting and Naming Cells and Ranges Specia Edition Using Microsoft Exce 2000 - Lesson 3 - Seecting and Naming Ces and.. Page 1 of 8 [Figures are not incuded in this sampe chapter] Specia Edition Using Microsoft Exce 2000-3 - Seecting and

More information

Area Efficient Implementation of Elliptic Curve Point Multiplication Algorithm

Area Efficient Implementation of Elliptic Curve Point Multiplication Algorithm (IJACSA) Internationa Journa of Advanced Coputer Science and Appications, Vo. 6, No., 5 Area Efficient Ipeentation of Eiptic Curve Point Mutipication Agorith Suni Devidas Bobade Research Schoar S.G.B.Aravati

More information

Analysing Real-Time Communications: Controller Area Network (CAN) *

Analysing Real-Time Communications: Controller Area Network (CAN) * Analysing Real-Tie Counications: Controller Area Network (CAN) * Abstract The increasing use of counication networks in tie critical applications presents engineers with fundaental probles with the deterination

More information

Oblivious Routing for Fat-Tree Based System Area Networks with Uncertain Traffic Demands

Oblivious Routing for Fat-Tree Based System Area Networks with Uncertain Traffic Demands Oblivious Routing for Fat-Tree Based Syste Area Networks with Uncertain Traffic Deands Xin Yuan Wickus Nienaber Zhenhai Duan Departent of Coputer Science Florida State University Tallahassee, FL 3306 {xyuan,nienaber,duan}@cs.fsu.edu

More information

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 13

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 13 Coputer Aided Drafting, Design and Manufacturing Volue 26, uber 2, June 2016, Page 13 CADDM 3D reconstruction of coplex curved objects fro line drawings Sun Yanling, Dong Lijun Institute of Mechanical

More information

A Trajectory Splitting Model for Efficient Spatio-Temporal Indexing

A Trajectory Splitting Model for Efficient Spatio-Temporal Indexing A Trajectory Splitting Model for Efficient Spatio-Teporal Indexing Slobodan Rasetic Jörg Sander Jaes Elding Mario A. Nasciento Departent of Coputing Science University of Alberta Edonton, Alberta, Canada

More information

Modeling Parallel Applications Performance on Heterogeneous Systems

Modeling Parallel Applications Performance on Heterogeneous Systems Modeling Parallel Applications Perforance on Heterogeneous Systes Jaeela Al-Jaroodi, Nader Mohaed, Hong Jiang and David Swanson Departent of Coputer Science and Engineering University of Nebraska Lincoln

More information

Advances in Crystallographic Image Processing for Scanning Probe Microscopy

Advances in Crystallographic Image Processing for Scanning Probe Microscopy Advances in Crystaographic Iage Processing for Scanning Probe Microscopy P. Moeck Nano-Crystaography Group, Departent of Physics, Portand State University, 1719 SW 10 th Avenue (SRTC), Portand, OR 97201,

More information

OPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS

OPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS Key words SOA, optial, coplex service, coposition, Quality of Service Piotr RYGIELSKI*, Paweł ŚWIĄTEK* OPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS One of the ost iportant tasks in service oriented

More information

Data & Knowledge Engineering

Data & Knowledge Engineering Data & Knowledge Engineering 7 (211) 17 187 Contents lists available at ScienceDirect Data & Knowledge Engineering journal hoepage: www.elsevier.co/locate/datak An approxiate duplicate eliination in RFID

More information

EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B.

EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B. EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B. Girod1 1 Stanford University, USA 2 Qualco Inc., USA ABSTRACT We study

More information

file://j:\macmillancomputerpublishing\chapters\in073.html 3/22/01

file://j:\macmillancomputerpublishing\chapters\in073.html 3/22/01 Page 1 of 15 Chapter 9 Chapter 9: Deveoping the Logica Data Mode The information requirements and business rues provide the information to produce the entities, attributes, and reationships in ogica mode.

More information

MULTI-INDEX VOTING FOR ASYMMETRIC DISTANCE COMPUTATION IN A LARGE-SCALE BINARY CODES. Chih-Yi Chiu, Yu-Cyuan Liou, and Sheng-Hao Chou

MULTI-INDEX VOTING FOR ASYMMETRIC DISTANCE COMPUTATION IN A LARGE-SCALE BINARY CODES. Chih-Yi Chiu, Yu-Cyuan Liou, and Sheng-Hao Chou MULTI-INDEX VOTING FOR ASYMMETRIC DISTANCE COMPUTATION IN A LARGE-SCALE BINARY CODES Chih-Yi Chiu, Yu-Cyuan Liou, and Sheng-Hao Chou Departent of Coputer Science and Inforation Engineering, National Chiayi

More information

Lecture outline Graphics and Interaction Scan Converting Polygons and Lines. Inside or outside a polygon? Scan conversion.

Lecture outline Graphics and Interaction Scan Converting Polygons and Lines. Inside or outside a polygon? Scan conversion. Lecture outine 433-324 Graphics and Interaction Scan Converting Poygons and Lines Department of Computer Science and Software Engineering The Introduction Scan conversion Scan-ine agorithm Edge coherence

More information

Ad Hoc Networks 11 (2013) Contents lists available at SciVerse ScienceDirect. Ad Hoc Networks

Ad Hoc Networks 11 (2013) Contents lists available at SciVerse ScienceDirect. Ad Hoc Networks Ad Hoc Networks (3) 683 698 Contents ists avaiabe at SciVerse ScienceDirect Ad Hoc Networks journa homepage: www.esevier.com/ocate/adhoc Dynamic agent-based hierarchica muticast for wireess mesh networks

More information

Data Management Updates

Data Management Updates Data Management Updates Jenny Darcy Data Management Aiance CRP Meeting, Thursday, November 1st, 2018 Presentation Objectives New staff Update on Ingres (JCCS) conversion project Fina IRB cosure at study

More information

A Learning Framework for Nearest Neighbor Search

A Learning Framework for Nearest Neighbor Search A Learning Fraework for Nearest Neighbor Search Lawrence Cayton Departent of Coputer Science University of California, San Diego lcayton@cs.ucsd.edu Sanjoy Dasgupta Departent of Coputer Science University

More information

1 Extended Boolean Model

1 Extended Boolean Model 1 EXTENDED BOOLEAN MODEL It has been well-known that the Boolean odel is too inflexible, requiring skilful use of Boolean operators to obtain good results. On the other hand, the vector space odel is flexible

More information

Geo-activity Recommendations by using Improved Feature Combination

Geo-activity Recommendations by using Improved Feature Combination Geo-activity Recoendations by using Iproved Feature Cobination Masoud Sattari Middle East Technical University Ankara, Turkey e76326@ceng.etu.edu.tr Murat Manguoglu Middle East Technical University Ankara,

More information

AgreeYa Solutions. Site Administrator for SharePoint User Guide

AgreeYa Solutions. Site Administrator for SharePoint User Guide AgreeYa Soutions Site Administrator for SharePoint 5.2.4 User Guide 2017 2017 AgreeYa Soutions Inc. A rights reserved. This product is protected by U.S. and internationa copyright and inteectua property

More information

Theoretical Analysis of Local Search and Simple Evolutionary Algorithms for the Generalized Travelling Salesperson Problem

Theoretical Analysis of Local Search and Simple Evolutionary Algorithms for the Generalized Travelling Salesperson Problem Theoretical Analysis of Local Search and Siple Evolutionary Algoriths for the Generalized Travelling Salesperson Proble Mojgan Pourhassan ojgan.pourhassan@adelaide.edu.au Optiisation and Logistics, The

More information

A CRYPTANALYTIC ATTACK ON RC4 STREAM CIPHER

A CRYPTANALYTIC ATTACK ON RC4 STREAM CIPHER A CRYPTANALYTIC ATTACK ON RC4 STREAM CIPHER VIOLETA TOMAŠEVIĆ, SLOBODAN BOJANIĆ 2 and OCTAVIO NIETO-TALADRIZ 2 The Mihajlo Pupin Institute, Volgina 5, 000 Belgrade, SERBIA AND MONTENEGRO 2 Technical University

More information

A Hybrid Network Architecture for File Transfers

A Hybrid Network Architecture for File Transfers JOURNAL OF IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 6, NO., JANUARY 9 A Hybrid Network Architecture for File Transfers Xiuduan Fang, Meber, IEEE, Malathi Veeraraghavan, Senior Meber,

More information

As Michi Henning and Steve Vinoski showed 1, calling a remote

As Michi Henning and Steve Vinoski showed 1, calling a remote Reducing CORBA Ca Latency by Caching and Prefetching Bernd Brügge and Christoph Vismeier Technische Universität München Method ca atency is a major probem in approaches based on object-oriented middeware

More information

Energy-Efficient Disk Replacement and File Placement Techniques for Mobile Systems with Hard Disks

Energy-Efficient Disk Replacement and File Placement Techniques for Mobile Systems with Hard Disks Energy-Efficient Disk Replaceent and File Placeent Techniques for Mobile Systes with Hard Disks Young-Jin Ki School of Coputer Science & Engineering Seoul National University Seoul 151-742, KOREA youngjk@davinci.snu.ac.kr

More information

Optimal Route Queries with Arbitrary Order Constraints

Optimal Route Queries with Arbitrary Order Constraints IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL.?, NO.?,? 20?? 1 Optial Route Queries with Arbitrary Order Constraints Jing Li, Yin Yang, Nikos Maoulis Abstract Given a set of spatial points DS,

More information

Defining and Surveying Wireless Link Virtualization and Wireless Network Virtualization

Defining and Surveying Wireless Link Virtualization and Wireless Network Virtualization 1 Defining and Surveying Wireless Link Virtualization and Wireless Network Virtualization Jonathan van de Belt, Haed Ahadi, and Linda E. Doyle The Centre for Future Networks and Counications - CONNECT,

More information

NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH

NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH V. Atienza; J.M. Valiente and G. Andreu Departaento de Ingeniería de Sisteas, Coputadores y Autoática Universidad Politécnica de Valencia.

More information

Modelling and Performance Evaluation of Router Transparent Web cache Mode

Modelling and Performance Evaluation of Router Transparent Web cache Mode Emad Hassan A-Hemiary IJCSET Juy 2012 Vo 2, Issue 7,1316-1320 Modeing and Performance Evauation of Transparent cache Mode Emad Hassan A-Hemiary Network Engineering Department, Coege of Information Engineering,

More information

QUERY ROUTING OPTIMIZATION IN SENSOR COMMUNICATION NETWORKS

QUERY ROUTING OPTIMIZATION IN SENSOR COMMUNICATION NETWORKS QUERY ROUTING OPTIMIZATION IN SENSOR COMMUNICATION NETWORKS Guofei Jiang and George Cybenko Institute for Security Technology Studies and Thayer School of Engineering Dartouth College, Hanover NH 03755

More information

Leveraging Relevance Cues for Improved Spoken Document Retrieval

Leveraging Relevance Cues for Improved Spoken Document Retrieval Leveraging Relevance Cues for Iproved Spoken Docuent Retrieval Pei-Ning Chen 1, Kuan-Yu Chen 2 and Berlin Chen 1 National Taiwan Noral University, Taiwan 1 Institute of Inforation Science, Acadeia Sinica,

More information

A Low-Cost Multi-Failure Resilient Replication Scheme for High Data Availability in Cloud Storage

A Low-Cost Multi-Failure Resilient Replication Scheme for High Data Availability in Cloud Storage 216 IEEE 23rd International Conference on High Perforance Coputing A Low-Cost Multi-Failure Resilient Replication Schee for High Data Availability in Cloud Storage Jinwei Liu* and Haiying Shen *Departent

More information

Enhancing Real-Time CAN Communications by the Prioritization of Urgent Messages at the Outgoing Queue

Enhancing Real-Time CAN Communications by the Prioritization of Urgent Messages at the Outgoing Queue Enhancing Real-Tie CAN Counications by the Prioritization of Urgent Messages at the Outgoing Queue ANTÓNIO J. PIRES (1), JOÃO P. SOUSA (), FRANCISCO VASQUES (3) 1,,3 Faculdade de Engenharia da Universidade

More information

A Novel Fast Constructive Algorithm for Neural Classifier

A Novel Fast Constructive Algorithm for Neural Classifier A Novel Fast Constructive Algorith for Neural Classifier Xudong Jiang Centre for Signal Processing, School of Electrical and Electronic Engineering Nanyang Technological University Nanyang Avenue, Singapore

More information

Joint Measurement- and Traffic Descriptor-based Admission Control at Real-Time Traffic Aggregation Points

Joint Measurement- and Traffic Descriptor-based Admission Control at Real-Time Traffic Aggregation Points Joint Measureent- and Traffic Descriptor-based Adission Control at Real-Tie Traffic Aggregation Points Stylianos Georgoulas, Panos Triintzios and George Pavlou Centre for Counication Systes Research, University

More information

Multi-level Shape Recognition based on Wavelet-Transform. Modulus Maxima

Multi-level Shape Recognition based on Wavelet-Transform. Modulus Maxima uti-eve Shape Recognition based on Waveet-Transform oduus axima Faouzi Aaya Cheikh, Azhar Quddus and oncef Gabbouj Tampere University of Technoogy (TUT), Signa Processing aboratory, P.O. Box 553, FIN-33101

More information

Arithmetic Coding. Prof. Ja-Ling Wu. Department of Computer Science and Information Engineering National Taiwan University

Arithmetic Coding. Prof. Ja-Ling Wu. Department of Computer Science and Information Engineering National Taiwan University Arithmetic Coding Prof. Ja-Ling Wu Department of Computer Science and Information Engineering Nationa Taiwan University F(X) Shannon-Fano-Eias Coding W..o.g. we can take X={,,,m}. Assume p()>0 for a. The

More information

A Comparison of a Second-Order versus a Fourth- Order Laplacian Operator in the Multigrid Algorithm

A Comparison of a Second-Order versus a Fourth- Order Laplacian Operator in the Multigrid Algorithm A Comparison of a Second-Order versus a Fourth- Order Lapacian Operator in the Mutigrid Agorithm Kaushik Datta (kdatta@cs.berkeey.edu Math Project May 9, 003 Abstract In this paper, the mutigrid agorithm

More information

A comparative study of colortexture

A comparative study of colortexture 12th Int. Worshop on Systes, Signas & Iage Processing, 22-24 Septeber 2005, haida, Greece 205 A coparative study of coortexture iage features Diitris Iaovidis* Departent of Inforatics and Teecounications,

More information

MAPPING THE DATA FLOW MODEL OF COMPUTATION INTO AN ENHANCED VON NEUMANN PROCESSOR * Peter M. Maurer

MAPPING THE DATA FLOW MODEL OF COMPUTATION INTO AN ENHANCED VON NEUMANN PROCESSOR * Peter M. Maurer MAPPING THE DATA FLOW MODEL OF COMPUTATION INTO AN ENHANCED VON NEUMANN PROCESSOR * Peter M. Maurer Departent of Coputer Science and Engineering University of South Florida Tapa, FL 33620 Abstract -- The

More information

DYNAMIC ESTIMATION OF BDP IN MANETS FOR EFFECTIVE NEXT NODE SELECTION

DYNAMIC ESTIMATION OF BDP IN MANETS FOR EFFECTIVE NEXT NODE SELECTION www.arpnjournals.co DYNAMIC ESTIMATION OF BDP IN MANETS FOR EFFECTIVE NEXT NODE SELECTION N. Snehalatha 1 and Paul Rodrigues 2 1 School of Coputing, SRM University, Chennai, Tail Nadu, India 2 Departent

More information

A Learning Framework for Nearest Neighbor Search

A Learning Framework for Nearest Neighbor Search A Learning Fraework for Nearest Neighbor Search Lawrence Cayton Departent of Coputer Science University of California, San Diego lcayton@cs.ucsd.edu Sanjoy Dasgupta Departent of Coputer Science University

More information

Feature Selection to Relate Words and Images

Feature Selection to Relate Words and Images The Open Inforation Systes Journal, 2009, 3, 9-13 9 Feature Selection to Relate Words and Iages Wei-Chao Lin 1 and Chih-Fong Tsai*,2 Open Access 1 Departent of Coputing, Engineering and Technology, University

More information

Image hiding with an improved genetic algorithm and an optimal pixel adjustment process

Image hiding with an improved genetic algorithm and an optimal pixel adjustment process Eighth International Conference on Intelligent Systes Design and Applications Iage hiding with an iproved genetic algorith and an optial pixel adjustent process Lin-Yu Tseng Yung-Kuan Chan Yu-An Ho Yen-Ping

More information

THE rapid growth and continuous change of the real

THE rapid growth and continuous change of the real IEEE TRANSACTIONS ON SERVICES COMPUTING, VOL. 8, NO. 1, JANUARY/FEBRUARY 2015 47 Designing High Perforance Web-Based Coputing Services to Proote Teleedicine Database Manageent Syste Isail Hababeh, Issa

More information

Real-Time Detection of Invisible Spreaders

Real-Time Detection of Invisible Spreaders Real-Tie Detection of Invisible Spreaders MyungKeun Yoon Shigang Chen Departent of Coputer & Inforation Science & Engineering University of Florida, Gainesville, FL 3, USA {yoon, sgchen}@cise.ufl.edu Abstract

More information

Self-Control Cyclic Access with Time Division - A MAC Proposal for The HFC System

Self-Control Cyclic Access with Time Division - A MAC Proposal for The HFC System Sef-Contro Cycic Access with Time Division - A MAC Proposa for The HFC System S.M. Jiang, Danny H.K. Tsang, Samue T. Chanson Hong Kong University of Science & Technoogy Cear Water Bay, Kowoon, Hong Kong

More information

HKBU Institutional Repository

HKBU Institutional Repository Hong Kong Baptist University HKBU Institutional Repository HKBU Staff Publication 18 Towards why-not spatial keyword top-k ueries: A direction-aware approach Lei Chen Hong Kong Baptist University, psuanle@gail.co

More information

Revisions for VISRAD

Revisions for VISRAD Revisions for VISRAD 16.0.0 Support has been added for the SLAC MEC target chamber: 4 beams have been added to the Laser System: X-ray beam (fixed in Port P 90-180), 2 movabe Nd:Gass (ong-puse) beams,

More information

TSR: Topology Reduction from Tree to Star Data Grids

TSR: Topology Reduction from Tree to Star Data Grids 03 Seventh Internationa Conference on Innovative Mobie and Internet Services in biquitous Computing TSR: Topoogy Reduction from Tree to Star Data Grids Ming-Chang Lee #, Fang-Yie Leu *, Ying-ping Chen

More information

AN EVOLUTIONARY APPROACH TO OPTIMIZATION OF A LAYOUT CHART

AN EVOLUTIONARY APPROACH TO OPTIMIZATION OF A LAYOUT CHART 13 AN EVOLUTIONARY APPROACH TO OPTIMIZATION OF A LAYOUT CHART Eva Vona University of Ostrava, 30th dubna st. 22, Ostrava, Czech Repubic e-mai: Eva.Vona@osu.cz Abstract: This artice presents the use of

More information

Guillotine subdivisions approximate polygonal subdivisions: Part III { Faster polynomial-time approximation schemes for

Guillotine subdivisions approximate polygonal subdivisions: Part III { Faster polynomial-time approximation schemes for Guillotine subdivisions approxiate polygonal subdivisions: Part III { Faster polynoial-tie approxiation schees for geoetric network optiization Joseph S. B. Mitchell y April 19, 1997; Last revision: May

More information

TALLINN UNIVERSITY OF TECHNOLOGY, INSTITUTE OF PHYSICS 17. FRESNEL DIFFRACTION ON A ROUND APERTURE

TALLINN UNIVERSITY OF TECHNOLOGY, INSTITUTE OF PHYSICS 17. FRESNEL DIFFRACTION ON A ROUND APERTURE 7. FRESNEL DIFFRACTION ON A ROUND APERTURE. Objective Exaining diffraction pattern on a round aperture, deterining wavelength of light source.. Equipent needed Optical workbench, light source, color filters,

More information

Clustering. Cluster Analysis of Microarray Data. Microarray Data for Clustering. Data for Clustering

Clustering. Cluster Analysis of Microarray Data. Microarray Data for Clustering. Data for Clustering Clustering Cluster Analysis of Microarray Data 4/3/009 Copyright 009 Dan Nettleton Group obects that are siilar to one another together in a cluster. Separate obects that are dissiilar fro each other into

More information

An Ad Hoc Adaptive Hashing Technique for Non-Uniformly Distributed IP Address Lookup in Computer Networks

An Ad Hoc Adaptive Hashing Technique for Non-Uniformly Distributed IP Address Lookup in Computer Networks An Ad Hoc Adaptive Hashing Technique for Non-Uniforly Distributed IP Address Lookup in Coputer Networks Christopher Martinez Departent of Electrical and Coputer Engineering The University of Texas at San

More information

TRANSFORMATIONS AND SYMMETRY

TRANSFORMATIONS AND SYMMETRY TRNSFORMTIONS ND SYMMETRY 1.2.1 1.2.5 Studing transforations of geoetric shapes buids a foundation for a ke idea in geoetr: congruence. In this introduction to transforations, the students epore three

More information

CSE120 Principles of Operating Systems. Prof Yuanyuan (YY) Zhou Advanced Memory Management

CSE120 Principles of Operating Systems. Prof Yuanyuan (YY) Zhou Advanced Memory Management CSE120 Principes of Operating Systems Prof Yuanyuan (YY) Zhou Advanced Memory Management Advanced Functionaity Now we re going to ook at some advanced functionaity that the OS can provide appications using

More information

MCSE Training Guide: Windows Architecture and Memory

MCSE Training Guide: Windows Architecture and Memory MCSE Training Guide: Windows 95 -- Ch 2 -- Architecture and Memory Page 1 of 13 MCSE Training Guide: Windows 95-2 - Architecture and Memory This chapter wi hep you prepare for the exam by covering the

More information

Formalising Refactorings with Graph Transformations

Formalising Refactorings with Graph Transformations Fundaenta Inforaticae XXI (200) 00 022 00 IOS Press Foraising Refactorings with Graph Transforations To Mens C Prograing Technoogy Lab Vrije Universiteit Brusse Peinaan 2, 050 Brusse, Begiu to.ens@vub.ac.be

More information

METAMORPHOSIS OF PERIODIC SURFACE MODELS

METAMORPHOSIS OF PERIODIC SURFACE MODELS Proceedings of ASE 2009 Internationa Design Engineering echnica Conferences & Coputers and Inforation in Engineering Conference IDEC/CIE 2009 August 0 Septeber 2, 2009, San Diego, Caifornia, USA DEC2009/DAC-870

More information

Quality of Service Evaluations of Multicast Streaming Protocols *

Quality of Service Evaluations of Multicast Streaming Protocols * Quaity of Service Evauations of Muticast Streaming Protocos Haonan Tan Derek L. Eager Mary. Vernon Hongfei Guo omputer Sciences Department University of Wisconsin-Madison, USA {haonan, vernon, guo}@cs.wisc.edu

More information

CSE120 Principles of Operating Systems. Prof Yuanyuan (YY) Zhou Scheduling

CSE120 Principles of Operating Systems. Prof Yuanyuan (YY) Zhou Scheduling CSE120 Principes of Operating Systems Prof Yuanyuan (YY) Zhou Scheduing Announcement Homework 2 due on October 25th Project 1 due on October 26th 2 CSE 120 Scheduing and Deadock Scheduing Overview In discussing

More information

Intro to Programming & C Why Program? 1.2 Computer Systems: Hardware and Software. Why Learn to Program?

Intro to Programming & C Why Program? 1.2 Computer Systems: Hardware and Software. Why Learn to Program? Intro to Programming & C++ Unit 1 Sections 1.1-3 and 2.1-10, 2.12-13, 2.15-17 CS 1428 Spring 2018 Ji Seaman 1.1 Why Program? Computer programmabe machine designed to foow instructions Program a set of

More information

Shortest Path Determination in a Wireless Packet Switch Network System in University of Calabar Using a Modified Dijkstra s Algorithm

Shortest Path Determination in a Wireless Packet Switch Network System in University of Calabar Using a Modified Dijkstra s Algorithm International Journal of Engineering and Technical Research (IJETR) ISSN: 31-869 (O) 454-4698 (P), Volue-5, Issue-1, May 16 Shortest Path Deterination in a Wireless Packet Switch Network Syste in University

More information

Trees. Linear vs. Branching CSE 143. Branching Structures in CS. What s in a Node? A Tree. [Chapter 10]

Trees. Linear vs. Branching CSE 143. Branching Structures in CS. What s in a Node? A Tree. [Chapter 10] CSE 143 Trees [Chapter 10] Linear vs. Branching Our data structures so far are linear Have a beginning and an end Everything falls in order between the ends Arrays, lined lists, queues, stacs, priority

More information

Novel Image Representation and Description Technique using Density Histogram of Feature Points

Novel Image Representation and Description Technique using Density Histogram of Feature Points Novel Iage Representation and Description Technique using Density Histogra of Feature Points Keneilwe ZUVA Departent of Coputer Science, University of Botswana, P/Bag 00704 UB, Gaborone, Botswana and Tranos

More information

Multi Packet Reception and Network Coding

Multi Packet Reception and Network Coding The 2010 Military Counications Conference - Unclassified Progra - etworking Protocols and Perforance Track Multi Packet Reception and etwork Coding Aran Rezaee Research Laboratory of Electronics Massachusetts

More information

Mapping Data in Peer-to-Peer Systems: Semantics and Algorithmic Issues

Mapping Data in Peer-to-Peer Systems: Semantics and Algorithmic Issues Mapping Data in Peer-to-Peer Systes: Seantics and Algorithic Issues Anastasios Keentsietsidis Marcelo Arenas Renée J. Miller Departent of Coputer Science University of Toronto {tasos,arenas,iller}@cs.toronto.edu

More information

Proceedings of the International Conference on Systolic Arrays, San Diego, California, U.S.A., May 25-27, 1988 AN EFFICIENT ASYNCHRONOUS MULTIPLIER!

Proceedings of the International Conference on Systolic Arrays, San Diego, California, U.S.A., May 25-27, 1988 AN EFFICIENT ASYNCHRONOUS MULTIPLIER! [1,2] have, in theory, revoutionized cryptography. Unfortunatey, athough offer many advantages over conventiona and authentication), such cock synchronization in this appication due to the arge operand

More information

CST-Trees: Cache Sensitive T-Trees

CST-Trees: Cache Sensitive T-Trees CST-Trees: Cache Sensitive T-Trees Ig-hoon Lee 1, Junho Shim 2, Sang-goo Lee 3, and Jonghoon Chun 4 1 Prompt Corp., Seoul, Korea ihlee@prompt.co.kr 2 Department of Computer Science, Sookmyung Women s University,

More information

Meeting Exchange 4.1 Service Pack 2 Release Notes for the S6200/S6800 Servers

Meeting Exchange 4.1 Service Pack 2 Release Notes for the S6200/S6800 Servers Meeting Exchange 4.1 Service Pack 2 Reease Notes for the S6200/S6800 Servers The Meeting Exchange S6200/S6800 Media Servers are SIP-based voice and web conferencing soutions that extend Avaya s conferencing

More information