Analysis of Dependencies of Checkpoint Cost and Checkpoint Interval of Fault Tolerant MPI Applications
|
|
- Myrtle Holmes
- 5 years ago
- Views:
Transcription
1 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng Analyss of Dependences of Checkpont Cost and Checkpont Interval of Fault Tolerant MPI Applcatons Mallkarjuna Shastry P.M.# 1 and K. Venkatesh # 2 #1 Sapthagr College of Engneerng, #2 M. S. Ramaah Insttute of Technology, # Afflated to Vshweshwaraya Technologcal Unversty, Bangalore, Karnataka, Inda., Abstract In ths paper, we have analysed ) the relatonshp between the checkpont cost and the optmal checkpont nterval and ) the relatonshp between the checkpont cost and the number of processors (processes) and we have also determned the optmal number of processors (processes) requred for executng the fault tolerant MPI applcatons. We have presented an expermental study n whch, we have used an optmal checkpont restart model [1] wth Webull s and Exponental dstrbutons to determne the optmal checkpont nterval. We have observed that, the optmal checkpont ntervals obtaned usng Webull s dstrbuton, produce mnmal average completon tme; as compared wth the optmal checkpont ntervals obtaned usng Exponental dstrbuton. The optmal checkpont nterval s approxmately drectly proportonal to the checkpont cost and nversely proportonal to shape parameter. The study ndcates that, the checkpont cost of MPI applcatons ncreases wth the number of processors (processes) used for executon. We have determned the optmal number of processes (processors) requred to execute the MPI applcatons consdered n ths paper, as 4. Keywords: OPEN MPI, Fault Tolerance, Optmal Checkpont Interval, Checkpont Cost, Speedup, Effcency. I. Introducton As the complexty of the program ncreases, the number of processors to be added to the cluster / HPC / Super Computer also ncreases, whch n turn decreases the MTBF (mean tme between falures) of the processors or the machnes [2]. When a processor fals or aborts before the completon of executon of the MPI applcaton, the MPI applcaton s restarted from the begnnng. Hence, t s requred to provde the fault tolerance to all the MPI applcatons whch run on multple processors or processes [1][3]-[7]. Fault tolerant MPI applcatons are checkponted perodcally and the checkponts are stored ether locally or sent to a remote server. When a fault tolerant MPI applcaton fals/aborts, t s restarted from the most recently saved checkpont on a local dsk. Hence, t s not requred to restart the applcaton from the begnnng [1][3]-[7]. The rollback recovery protocols such as uncoordnated, coordnated, communcaton nduced checkpontng and message loggng protocols [5] can be used to acheve the fault tolerance. We have used the coordnated checkpontng protocol [5][7] to provde the fault tolerance to MPI applcatons consdered n ths paper. MPI collectve communcatons are used to develop the MPI applcatons [7]. The optmal checkpont restart model (OCRM) developed by M. Shastry et al. [1] has been used n ths paper, to determne the optmal checkpont nterval, when the scale parameter and shape parameter of Webull s dstrbuton are known along wth checkpont cost. Fxed checkpont nterval s used to checkpont the MPI applcatons consdered n ths paper [8]. The checkpontng of MPI applcatons ncludes dfferent costs lke, checkpont cost, rollback cost, and restart cost [1][8][9]. These costs are collectvely called the checkpont overheads. The average completon tme of an MPI applcaton s obtaned by addng the checkpont overheads to the executon tme of MPI applcaton [1][8][9]. We would lke to determne ) the optmal number of processes (processors) requred to execute the MPI applcatons of dfferent sze, so that the average completon tme of these MPI applcatons wth checkpontng would be mnmal, ) the relatonshp between the checkpont cost and the optmal checkpont nterval, and ) the relatonshp between checkpont cost and number of processes used to execute the MPI applcatons consdered n ths paper. ISSN :
2 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng Rest of the paper s organzed as follows. In secton 2, the related works carred out on the study of relatonshp between checkpont cost and checkpont nterval, checkpont / restart models used to determne checkpont ntervals are dscussed. In secton 3, the dfferent parameters used and the assumptons made n ths paper are presented. In secton 4, determnaton of the optmal checkpont nterval usng Webull s and Exponental dstrbutons s dscussed brefly. In secton 5, the equatons used for determnng the average completon tme of the MPI applcatons are presented. In secton 6, the results of the case studes consdered n ths paper are presented. In secton 7, the expermental setup used n the analyss s dscussed and n secton 8, conclusons are presented. II. Related Works An excellent survey on rollback recovery protocols has been carred out by Elnozahy et al.[5]. Chandy[10][11] and Treaster[12] have presented a survey on rollback and recovery strateges used n fault tolerant MPI applcatons. M. Shastry et al. [8] have shown that the fxed checkpont nterval reduces the checkpont overheads of fault tolerant MPI applcatons as compared wth the ncremental or varyng checkpont nterval. Hence, we have used the fxed checkpont nterval n our experment to checkpont the MPI applcatons. Geff et al. [9] have used a smulator to determne the relatonshp between the checkpont nterval and the varous checkpont overheads. The OCRM dscussed by M.Shastry et al. [1] have shown the sgnfcant mprovement wth regard to the total wastage tme as compared wth the checkpont restart models developed by Yudun lu et al. [13][14] and Bouguerra Mohammed Slm et al. [15], The checkpont nterval obtaned by OCRM [1] usng Webull s and Exponental dstrbutons, produces mnmal checkpont overheads as compared wth Yudun lu et al. [13][14] and Bouguerra Mohammed Slm et al. [15]. Hoda El-Sayed et al. [16] and Xn Wang [17] have dscussed about the speedup and the effcency of MPI applcatons. III. Parameters/ Notatons Used and Assumptons Made. A. Parameters and Notatons Used. The parameters/notatons used n ths paper are presented n the table 1. Parameter /Notaton T C T S β α T N RB CC R TL F TL ET ACT T 1 T P Mat4k Mat5k Prme4L Prme5L Table 1. Parameters/Notatons used. Meanng Optmal Checkpont nterval. Tme requred to save the checkpont on a local dsk (Checkpont Cost). Shape Parameter of Webull s dstrbuton. Scale Parameter of Webull s dstrbuton. Executon tme tll a falure occurs n th cycle. Number of checkponts taken tll a falure occurs n th cycle. Rollback cost due to a falure n th cycle Checkpont cost n th cycle Restart cost(tme requred to resume the executon of the applcaton after a falure). Total tme lost n th cycle. Number of falures. Total tme lost due to F falures durng the executon of MPI applcaton. Executon Tme of MPI applcaton wthout checkpontng. Average Completon Tme of MPI Applcaton. Tme requred to execute the applcaton sequentally (usng a sngle processor) wthout checkpontng. Tme requred for parallel executon of the same applcaton usng p processors wthout checkpontng. MPI Applcaton, whch multples two matrces of sze 4000 * 4000 ntegers. MPI Applcaton, whch multples two matrces of sze 5000 * 5000 ntegers. MPI Applcaton, whch generates 4 lakh prme numbers on each process(processor). MPI Applcaton, whch generates 5 lakh prme numbers on each process (processor). ISSN :
3 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng B. Assumptons Made We have made the followng assumptons, whch are smlar to the assumptons made n [1] to checkpont the MPI applcatons consdered n ths paper. 1. A seres of random falures F ( = 1, 2, 3,..N) may nterrupt the executon of MPI applcaton. 2. A separate montorng software system s used to montor contnuously the falure of a fault tolerant MPI applcaton. 3. Checkpont nterval (T C ) s fxed and a checkpont s taken perodcally after the tme T C. 4. When a falure occurs durng the executon of MPI applcaton, MPI applcaton s rolled back to the most recent checkpont usng locally stored checkpont fle. 5. Tme requred for wrtng a checkpont (T S ) onto a local dsk s a constant and only one copy of the most recent checkpont s stored on a local dsk. 6. Tme requred for resumng/restartng (restart cost, R) the MPI applcaton from the most recent checkpont s a constant. IV. Determnaton of Optmal Checkpont Interval A. Usng Webull s Dstrbuton The optmal checkpont nterval T C s obtaned by usng the optmal checkpont restart model (OCRM) developed by [1] usng Webull s dstrbuton. The algorthm Estmate-T C ( ) of [1] s used wth the shape parameter β and the scale parameter α for the gven checkpont cost, T S. The shape parameter β s set to 0.5. B. Usng Exponental Dstrbuton. The optmal checkpont nterval T C s obtaned by usng the optmal checkpont restart model (OCRM) developed by [1] usng Exponental dstrbuton. The algorthm Estmate- T C ( ) of [1] s used wth the shape parameter β and the scale parameter α for the gven checkpont cost, T S. The shape parameter β s set to 1.0. V. Determnaton of Total Tme Lost The number of checkponts to be taken n a cycle before the occurrence of a falure can be determned by [1][8] N T ( T T ) (1 / C S ) Then, the cost of checkpont n th cycle s computed as follows [1][8]. CC N T S (2) The cost of rollback n th cycle s then computed as follows [1][8]. Rb ( T N ( T C T S )) (3) The tme lost n th cycle TL due to a falure can be obtaned by addng checkpont cost, rollback cost and restart cost together as follows[1][8]. TL CC Rb R (4) The tme lost n th cycle can also be computed [1] as follows by substtutng (2) and (3) n (4). TL T N T C R (5) The total tme lost durng the executon of a fault tolerant MPI applcaton s determned usng the followng equaton [1][8]. TL F 1 TL The average completon tme of a MPI applcaton s computed as follows ACT ET TL VI. Case Studes (6 ) (7 ) A. Varaton n Executon Tme (ET) wth Number of Processors (Processes). Varaton n executon tme of Mat4k and Mat5k wth number of processes s represented n fgure 1 a). Fgure 1a) shows that, the executon tme of Mat4k and Mat5k s very hgh, when number of processes s 1. When number of processes s ncreased to 2, the executon tme of both applcatons s reduced to almost 50%. When the number of processes s ncreased to 4 and 8, the executon tme of Mat5k ncreases lnearly, but, the executon tme of Mat4k remans almost constant. When the number of processes s ncreased to 16 and 32, the executon tme remans almost constant for both applcatons as shown n fgure 1a). Varaton n executon tme of Prme4L and Prme5L wth number of processes s represented n fgure 1 b). Fgure 1b) shows that, the executon tme of Prme4L and Prme5L s very hgh when number of processes s 1. When number of processes s ncreased to 2, 4 and 8 the executon tme of both applcatons s reduced by 26% to 40%. The executon tme of both applcatons remans almost constant, when the number of processes s ncreased to 16 and 32 as shown n fgure 1b). ISSN :
4 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng When P processors (processes) are used to execute the MPI applcatons, the speed up and the effcency [16][17] are computed as follows. Speed up = T 1 / T P. Effcency = Speed up / P. Varaton n speed up wth the number of processes used for executng the MPI applcatons consdered n ths paper, s represented n the fgure 1c). The fgure 1c) shows that, when the number of processes s between 1 and 4, the speed up s between 1 and 2 n all the 4 cases. Speedup remans same n case of Mat4k and Mat5k, when the number of processes s between 1 and 32. But the speedup ncreases gradually n case of Prme4L and Prme5L, when the number of processes s between 1 and 16 and remans constant, when the number of processes s between 16 and 32. It s clear from the fgure 1c) that, the speedup remans same n all the four cases, when the number of processes s 4. Varaton n effcency wth the number of processes used for executng the MPI applcatons consdered n ths paper, s represented n the fgure 1d). The fgure 1d) shows that, when the number of processes s between 1 and 4, effcency s between 0.5 and 1 n all the 4 cases. Effcency reduces gradually n all the 4 cases, as the number of processes s ncreased from 4 to 16 and reaches almost zero, when number of processes s 32. When the number of processors (processes) used for executng the MPI applcatons s P, speedup s above 2 and the effcency s between 0.5 and 1, the optmal number of processors (processes), P O, to be used for executng MPI applcatons s, P O = P. Snce the speedup s 2 and the effcency s above 0.5 as shown n fgures 1c) and 1d) respectvely, n all the 4 cases, we conclude that, the optmal number of processors (processes) to be used for executng all the 4 MPI applcatons consdered n ths paper, s, 4 (P O = 4). Fg 1 b). Varaton n Executon Tme of Prme4L and Prme5L. Fg 1c). Varaton n Speedup. Fg 1 a). Varaton n Executon Tme of Mat4k and Mat5k. Fg 1d). Varaton n Effcency. ISSN :
5 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng B. Varaton n Checkpont Cost wth Number of Processes (Processors). Fgures 2a) and 2b) present the varaton n checkpont cost of Mat4K, Mat5K and Prme4L, Prme5L MPI applcatons wth the number of processes respectvely. It s clear from the fgures 2a) and 2b) that, the checkpont cost of MPI applcatons ncreases lnearly as the number of processes ncreases from 1 to 32. Thus, the checkpont cost s drectly proportonal to the number of processes used for executng the MPI applcatons. Fgure 3 represents the optmal checkpont ntervals obtaned from OCRM [1] usng Webull s dstrbuton wth shape parameter, β = 0.5 and Exponental dstrbuton wth shape parameter, β = 1.0 for the dfferent values of checkpont cost, T S of MPI applcatons consdered n ths paper. The scale parameter α, s set to 4.8 mnutes (λ = 300 falures per day). In fgure 3, when β = 0.5, the T C remans 1.5 mnutes, when T S s less than 8 seconds and T C ncreases to 2 mnutes, when T S s between 9 and 16 seconds. The T C value ncreases gradually when T S s between 16 and 30 seconds. The value of T C decreases gradually further when T S s above 30 seconds. When β = 1.0, the optmal T C remans 0.5 mnutes, when T S s less than 5 seconds and T C ncreases to 1 mnute when T S s between 5 and 6 seconds. The T C value ncreases gradually when T S s between 10 and 42 seconds. The value of T C remans almost constant, when T S s above 42 seconds. Fg 2 a). Varaton n Checkpont Cost of Mat4k and Mat5k. Thus, t s clear from the fgure 3 that, the optmal checkpont ntervals obtaned from OCRM [1] usng Exponental dstrbuton wth β = 1.0 are low as compared to the optmal checkpont ntervals obtaned from OCRM [1] usng Webull s dstrbuton wth β = 0.5. Ths clearly ndcates that, the number of checkponts to be taken for an MPI applcaton s more n the case of Exponental dstrbuton (β = 1.0) as compared wth the Webull s dstrbuton(β = 0.5). It s also clear from the fgure 3 that, the optmal checkpont nterval s approxmately drectly proportonal to the checkpont cost, T S. of MPI applcaton and nversely proportonal to the shape parameter, β of Webull s dstrbuton. Fg 2 b). Varaton n Checkpont Cost of Prme4L and Prme5L. C. Varaton n Optmal Checkpont Interval wth Checkpont Cost. Fg 3. Varaton n Optmal Checkpont Interval ISSN :
6 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng D. Varaton n Average Completon tme wth Number of Processors (Processes). Fgures 4 and 5 represent the varaton n the average completon tme of Mat4k and Mat5k wth number of processes respectvely. Fgures 6 and 7 represent the varaton n average completon tme of Prme4L and Prme5L wth number of processes, respectvely. The average completon tme of all 4 MPI applcatons s computed usng the equaton (7). The average completon tme of Mat4k, as shown n the fgure 4, s almost same for β = 0.5 and β = 1.0, when the number of processes s between 1 and 16. Smlarly, the average completon tme of Mat5k, as shown n fgure 5, s almost same for β = 0.5 and β = 1.0, when the number of processes s between 1 and 16 Snce, the optmal checkpont ntervals obtaned from OCRM [1] usng exponental dstrbuton wth β = 1.0 are low as compared wth the optmal checkpont ntervals obtaned from OCRM [1] usng Webull s dstrbuton wth β = 0.5 as shown n fgure 3, the average completon tme of Mat4k s 9% less and Mat5k s 7% less, when β = 0.5, as shown n the fgures 4 and 5 respectvely, when the number of processes s above 16. Ths result s as expected from OCRM[1], because, the number of checkponts to be taken wll be more n the case of Exponental dstrbuton wth β = 1.0 as compared to the Webull s dstrbuton wth β = 0.5. Fg 4. Average Completon Tme of Mat4k The average completon tme of the Prme4L, as shown n fgure 6, s almost same for both β = 0.5 and β = 1.0, when the number of processes s between 1 and 32. Smlarly, the average completon tme of the Prme5L, as shown n fgure 7, s almost same for both β = 0.5 and β = 1.0, when the number of processes s between 1 and 32. But, stll the average completon tme of these applcatons s 3 % less, when β = 0.5, as shown n fgures 6 and 7. Thus, t s clear from the fgures 4, 5, 6 and 7 that, the optmal checkpont ntervals obtaned from OCRM [1] usng Webull s dstrbuton wth β = 0.5, yeld mnmal average completon tme as compared to the optmal checkpont ntervals obtaned from OCRM [1] usng Exponental dstrbuton wth β = 1.0, rrespectve of the number of processes used to execute the MPI applcatons consdered n ths paper. Fg 5. Average Completon Tme of Mat5k ISSN :
7 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng applcatons. BLCR s a LINUX based coordnated checkpontng protocol [19]. We have used fxed checkpont nterval to checkpont the MPI applcatons [8], as t reduces the total waste tme due to the checkpontng of MPI applcatons. The above mentoned MPI applcatons are executed and checkponted perodcally after the tme T C. We have generated λ = 300 falures per day (α = 4.8 mnutes) durng the executon of above MPI applcatons. Webull s dstrbuton wth β = 0.5 and Exponental dstrbuton wth β = 1.0 are used to determne the optmal checkpont ntervals usng OCRM [1], when α = 4.8 mnutes. Fg 6. Average Completon Tme of Prme4L The montor program s developed n a shell scrpt. Ths program runs at the background and montors contnuously the executon of the MPI applcatons. Once, the montor program learns that the MPI applcaton has faled due to an nterrupton; t uses the most recent checkpont stored locally and resumes the executon of the MPI applcaton from the most recent checkpont. VIII. Conclusons. Executon tme of the MPI applcatons as shown n fgures 1a) and 1b), reduces to more than 30% to 50% when the number of processes s ncreased from 1 to 8 and the executon tme almost remans constant when the number of processes s ncreased from 8 to 32. Snce, when the number of processes s 4, for all the 4 applcatons, the speedup s 2 and the effcency s above 0.5 as shown n fgures 1c) and 1d) respectvely, we conclude that, the optmal number of processes to be used for executng all the 4 MPI applcatons consdered n ths paper, s 4. Fg 7. Average Completon Tme of Prme5L VII. Expermental Setup An AMD Athlon II X (Quad Core) processor wth 4 GB of RAM s used to execute the MPI applcatons. We have mplemented 4 MPI applcatons, Mat4k, Mat5k, Prme4L and Prme5L n C programmng by usng OPEN MPI [18] on a standalone system. OPEN MPI [18] allows the executon and checkpontng of MPI applcatons on standalone systems also. MPI collectve communcaton methods lke MPI_Scatter() and MPI_Bcast() are used to dstrbute the data to dfferent processes nvolved n the cluster[7]. Berkeley Lab s Checkpont restart (BLCR) technque along wth OPEN MPI s used to checkpont the above MPI Fgures 2a) and 2b) show that, the checkpont cost of MPI applcatons ncreases gradually as the number of processes ncreases from 1 to 32. Thus, the checkpont cost s drectly proportonal to the number of processes used to execute an MPI applcaton. Fgure 3 shows that, the optmal checkpont ntervals obtaned usng Exponental dstrbuton wth β = 1.0 are low as compared to the optmal checkpont ntervals obtaned usng Webull s dstrbuton wth β = 0.5. Ths clearly ndcates that, the number of checkponts to be taken for an MPI applcaton s more n the case of Exponental dstrbuton (β = 1.0) as compared wth the Webull s dstrbuton (β = 0.5). It s also clear from the fgure 3 that, the optmal checkpont nterval s approxmately drectly proportonal to the checkpont cost, T S. of MPI applcaton and nversely proportonal the shape parameter, β of Webull s dstrbuton. ISSN :
8 Mallkarjuna Shastry P M et al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng It s qute clear from the fgures 4, 5, 6 and 7 that, the optmal checkpont ntervals obtaned from OCRM [1] usng Webull s dstrbuton wth β = 0.5, yeld mnmal average completon tme as compared wth the optmal checkpont ntervals obtaned from [1] usng Exponental dstrbuton wth β = 1.0. Ths s because; the number of checkponts to be taken s more n case of Exponental dstrbuton as compared wth Webull s dstrbuton. Our results dscussed n ths paper agan emphass that, when the scale parameter α and the checkpont cost, T S of MPI applcatons are known, the Webull s dstrbuton can be used to determne optmal checkpont ntervals from OCRM [1] for any gven value of shape parameter, β. REFERENCES [1] Mallkarjuna Shastry P.M and K.Venkatesh, An Optmal Checkpont Restart Model usng Webull s and Exponental dstrbutons for Fault Tolerant Large Scale MPI applcatons, Beng processed by JPDC, Elsever, Un Publshed. [2] A.R. Adga, G.almas, and et al., An Overvew of the BlueGene/L Supercomputer, In Proceedngs of Supercomputng, IEEE/ACM Conference, pp60-60,2002. [3] Lus Moura Slva and Joao Gabrel Slva, The Performance of Coordnated and Independent Checkpontng, IEEE Trans, [4] G.E. Fagg, A. Bukovsky and J.J. Dongarra, Harness and Fault Tolerant MPI, Parallel Computng, 27(11): , [5] E. N. Elnozahy, Lorenzo Alvs, Y-Mn Wang, and B. Davd Johnson, A Survey of Rollback-Recovery Protocols n Message- Passng Systems, ACM Computng Surveys, Vol. 34, No. 3, pp , Sep [6] M.Treaster, A survey of fault-tolerance and fault-recovery technques n parallel systems, Techncal Report cs.dc / , ACM computng Research Repostory (CoRR), January [7] Mallkarjuna Shastry P.M, K.Venkatesh, Performance Evaluaton of Coordnated Checkpontng Protocol usng MPI pont to pont and collectve communcatons, IJACC, Accepted, n Press. [8] Mallkarjuna Shastry P.M. and K.Venkatesh, Selecton of a Checkpont Interval n Coordnated Checkpontng Protocol for Fault Tolerant Open MPI, IJCSE, Vol. 02, No. 06, pp , Sep [9] Geff Vallee, Anand Tkotekar, Stephen L Scott, Impact of Fault Tolerant Polces: Feasblty Study, Oak Natonal Laboratory, Apr [10] K.M. Chandy, A survey of analytc models of roll-back and recovery strateges, Computer 8, 5 (May 1975), [11] K.M. Chandy, J.C. Browne, C. W. Dssly, and W. R. Uhrg, Analytc models for rollback and recovery stratagems n data base systems, IEEE Trans Software Engg. SE-1, ( March 1975), [12] M.Treaster, A survey of fault-tolerance and fault-recovery technques n parallel systems, Techncal Report cs.dc / , ACM computng Research Repostory (CoRR), January [13] Y. Lu, Relablty Aware Optmal Checkpont/ Restart Model n Hgh Performance Computng, PhD Thess, Lousana Tech unversty, Ruston, LA, USA, May [14] Yudun Lu, Raja Nassar, Chokcha (box) Leangsuksum, Nchamon Naksnehaboon, Mhaels Paun, Stephen L. Scott, An Optmal Checkpont /Restart Model for a Large Scale Hgh Performance Computng System, IEEE Trans [15] Bouguerra Mohammed Slm, Therry Gauter, Dens Trystram, Jean,Marc Vncent, A New flexble Checkpont/Restart Model, INRIA, Dec [16] Hoda El-Sayed and Erc Wrght, Sparse Matrx Multplcaton Usng UPC, Proceedngs of the World Congress on Engneerng 2007, Vol. I, WCE 2007, July 2-4, 2007, London, U.K. [17] Xn Wang, Scalable Parallel Matrx MultplcatonAlgorthms wth Applcaton to a Maxmum Entropy Problem, Sep [18] The Open MPI Team, Open MPI Checkpont/Restart Fault Tolerance User s Gude, OPEN MPI Team, Oct [19] H. Paul Hargrove and C. Jason Duell, Berkeley lab checkpont / restart (BLCR) for Lnux clusters, Journal of Physcs, Conference seres 46 (2006), , ScDAC AUTHORS PROFILE MR. Mallkarjuna Shastry P.M has receved hs B.E. and M.Tech n Computer Scence and Engneerng from Karnataka Unversty Dharwar, and Vshweshwaraah Technologcal Unversty, Belgaum, Karnataka, Inda respectvely. He s currently pursung Ph.D on Analyss of Fault Tolerant Methods and Performance Evaluaton n Dstrbuted Systems under the gudance of Prof. Dr. K.Venkatesh at M.S.Ramaah Insttute of Technology, Bangalore-54, Karnataka, Inda. He s a Professor n the department of Computer Scence and Engg. at Sapthagr College of Engg, Bangalore, Karnataka, Inda. He has totally 18 years of teachng experence n Computer Scence and Engg. Prof. Dr. K. Venkatesh has receved hs M.Sc. n Physcs from Mysore Unversty n 1973 and MS from BITS Plan n 2001 respectvely. He has receved Ph.D n 1980 from Mysore Unversty. He s a professor at M.S.Ramaah Insttute of Technology, Bangalore-54 and has totally 30 years of teachng experence. ISSN :
Selection of a Checkpoint Interval in Coordinated Checkpointing Protocol for Fault Tolerant Open MPI
Mallkarjuna hastry et. al. / (IJE) Internatonal Journal on omputer cence and Engneerng Vol. 2, No. 6, 21, 264-27 electon of a heckpont Interval n oordnated heckpontng Protocol for Fault olerant Open MPI
More informationParallelism for Nested Loops with Non-uniform and Flow Dependences
Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr
More informationTerm Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task
Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto
More informationSimulation Based Analysis of FAST TCP using OMNET++
Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationProblem Definitions and Evaluation Criteria for Computational Expensive Optimization
Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty
More informationSkew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach
Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research
More informationAn Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices
Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal
More informationLearning the Kernel Parameters in Kernel Minimum Distance Classifier
Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department
More informationX- Chart Using ANOM Approach
ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are
More informationVirtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory
Background EECS. Operatng System Fundamentals No. Vrtual Memory Prof. Hu Jang Department of Electrcal Engneerng and Computer Scence, York Unversty Memory-management methods normally requres the entre process
More informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationAn Optimal Algorithm for Prufer Codes *
J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,
More informationMaintaining temporal validity of real-time data on non-continuously executing resources
Mantanng temporal valdty of real-tme data on non-contnuously executng resources Tan Ba, Hong Lu and Juan Yang Hunan Insttute of Scence and Technology, College of Computer Scence, 44, Yueyang, Chna Wuhan
More informationA mathematical programming approach to the analysis, design and scheduling of offshore oilfields
17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and
More informationSolving two-person zero-sum game by Matlab
Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by
More informationReducing Frame Rate for Object Tracking
Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg
More informationImprovement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration
Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,
More informationFor instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)
Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A
More informationFault tolerant workflow scheduling based on replication and resubmission of tasks in Cloud Computing
Fault tolerant workflow schedulng based on replcaton and resubmsson of tasks n Cloud Computng Jayadvya S K* Department of Computer Scence & Engneerng Natonal Insttute of Technology Truchrappall, Taml Nadu,
More informationComparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments
Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,
More informationPositive Semi-definite Programming Localization in Wireless Sensor Networks
Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer
More informationContent Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers
IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth
More informationTransaction-Consistent Global Checkpoints in a Distributed Database System
Proceedngs of the World Congress on Engneerng 2008 Vol I Transacton-Consstent Global Checkponts n a Dstrbuted Database System Jang Wu, D. Manvannan and Bhavan Thurasngham Abstract Checkpontng and rollback
More informationDetermining the Optimal Bandwidth Based on Multi-criterion Fusion
Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn
More informationEvaluation of Parallel Processing Systems through Queuing Model
ISSN 2278-309 Vkas Shnde, Internatonal Journal of Advanced Volume Trends 4, n Computer No.2, March Scence - and Aprl Engneerng, 205 4(2), March - Aprl 205, 36-43 Internatonal Journal of Advanced Trends
More informationSum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints
Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan
More informationUB at GeoCLEF Department of Geography Abstract
UB at GeoCLEF 2006 Mguel E. Ruz (1), Stuart Shapro (2), June Abbas (1), Slva B. Southwck (1) and Davd Mark (3) State Unversty of New York at Buffalo (1) Department of Lbrary and Informaton Studes (2) Department
More informationVirtual Machine Migration based on Trust Measurement of Computer Node
Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on
More informationTsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance
Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for
More informationConcurrent Apriori Data Mining Algorithms
Concurrent Apror Data Mnng Algorthms Vassl Halatchev Department of Electrcal Engneerng and Computer Scence York Unversty, Toronto October 8, 2015 Outlne Why t s mportant Introducton to Assocaton Rule Mnng
More informationReal-time Fault-tolerant Scheduling Algorithm for Distributed Computing Systems
Real-tme Fault-tolerant Schedulng Algorthm for Dstrbuted Computng Systems Yun Lng, Y Ouyang College of Computer Scence and Informaton Engneerng Zheang Gongshang Unversty Postal code: 310018 P.R.CHINA {ylng,
More informationLobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide
Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.
More informationNUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS
ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana
More informationAn Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method
Internatonal Journal of Computatonal and Appled Mathematcs. ISSN 89-4966 Volume, Number (07), pp. 33-4 Research Inda Publcatons http://www.rpublcaton.com An Accurate Evaluaton of Integrals n Convex and
More informationA Robust Method for Estimating the Fundamental Matrix
Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.
More informationRelated-Mode Attacks on CTR Encryption Mode
Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory
More informationCompiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz
Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster
More informationAn Efficient Garbage Collection for Flash Memory-Based Virtual Memory Systems
S. J and D. Shn: An Effcent Garbage Collecton for Flash Memory-Based Vrtual Memory Systems 2355 An Effcent Garbage Collecton for Flash Memory-Based Vrtual Memory Systems Seunggu J and Dongkun Shn, Member,
More informationClassifier Selection Based on Data Complexity Measures *
Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.
More informationProper Choice of Data Used for the Estimation of Datum Transformation Parameters
Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and
More informationNon-Split Restrained Dominating Set of an Interval Graph Using an Algorithm
Internatonal Journal of Advancements n Research & Technology, Volume, Issue, July- ISS - on-splt Restraned Domnatng Set of an Interval Graph Usng an Algorthm ABSTRACT Dr.A.Sudhakaraah *, E. Gnana Deepka,
More informationSpace-Optimal, Wait-Free Real-Time Synchronization
1 Space-Optmal, Wat-Free Real-Tme Synchronzaton Hyeonjoong Cho, Bnoy Ravndran ECE Dept., Vrgna Tech Blacksburg, VA 24061, USA {hjcho,bnoy}@vt.edu E. Douglas Jensen The MITRE Corporaton Bedford, MA 01730,
More informationA Progressive Fault Tolerant Mechanism in Mobile Agent Systems
A Progressve Fault Tolerant Mechansm n Moble Agent Systems Mchael R. LYU and Tsz Yeung WONG Computer Scence & Engneerng Department The Chnese Unversty of Hong Kong Shatn, Hong Kong flyu,tywongg@cse.cuhk.edu.hk
More informationAccounting for the Use of Different Length Scale Factors in x, y and z Directions
1 Accountng for the Use of Dfferent Length Scale Factors n x, y and z Drectons Taha Soch (taha.soch@kcl.ac.uk) Imagng Scences & Bomedcal Engneerng, Kng s College London, The Rayne Insttute, St Thomas Hosptal,
More informationAnalysis of Continuous Beams in General
Analyss of Contnuous Beams n General Contnuous beams consdered here are prsmatc, rgdly connected to each beam segment and supported at varous ponts along the beam. onts are selected at ponts of support,
More informationMathematics 256 a course in differential equations for engineering students
Mathematcs 56 a course n dfferental equatons for engneerng students Chapter 5. More effcent methods of numercal soluton Euler s method s qute neffcent. Because the error s essentally proportonal to the
More informationParallel matrix-vector multiplication
Appendx A Parallel matrx-vector multplcaton The reduced transton matrx of the three-dmensonal cage model for gel electrophoress, descrbed n secton 3.2, becomes excessvely large for polymer lengths more
More informationNetwork Coding as a Dynamical System
Network Codng as a Dynamcal System Narayan B. Mandayam IEEE Dstngushed Lecture (jont work wth Dan Zhang and a Su) Department of Electrcal and Computer Engneerng Rutgers Unversty Outlne. Introducton 2.
More informationS1 Note. Basis functions.
S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type
More informationConstructing Minimum Connected Dominating Set: Algorithmic approach
Constructng Mnmum Connected Domnatng Set: Algorthmc approach G.N. Puroht and Usha Sharma Centre for Mathematcal Scences, Banasthal Unversty, Rajasthan 304022 usha.sharma94@yahoo.com Abstract: Connected
More informationEvaluation of an Enhanced Scheme for High-level Nested Network Mobility
IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.15 No.10, October 2015 1 Evaluaton of an Enhanced Scheme for Hgh-level Nested Network Moblty Mohammed Babker Al Mohammed, Asha Hassan.
More informationModule Management Tool in Software Development Organizations
Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,
More informationQuery Clustering Using a Hybrid Query Similarity Measure
Query clusterng usng a hybrd query smlarty measure Fu. L., Goh, D.H., & Foo, S. (2004). WSEAS Transacton on Computers, 3(3), 700-705. Query Clusterng Usng a Hybrd Query Smlarty Measure Ln Fu, Don Hoe-Lan
More informationA MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS
Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung
More informationSLAM Summer School 2006 Practical 2: SLAM using Monocular Vision
SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,
More informationCordial and 3-Equitable Labeling for Some Star Related Graphs
Internatonal Mathematcal Forum, 4, 009, no. 31, 1543-1553 Cordal and 3-Equtable Labelng for Some Star Related Graphs S. K. Vadya Department of Mathematcs, Saurashtra Unversty Rajkot - 360005, Gujarat,
More informationLoad-Balanced Anycast Routing
Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance
More informationA Concurrent Non-Recursive Textured Algorithm for Distributed Multi-Utility State Estimation
1 A Concurrent Non-ecursve Textured Algorthm for Dstrbuted Mult-Utlty State Estmaton Garng M. Huang, Senor Member, IEEE, and Jansheng Le, Student Member, IEEE Abstract: Durng power deregulaton, power companes
More informationA Background Subtraction for a Vision-based User Interface *
A Background Subtracton for a Vson-based User Interface * Dongpyo Hong and Woontack Woo KJIST U-VR Lab. {dhon wwoo}@kjst.ac.kr Abstract In ths paper, we propose a robust and effcent background subtracton
More informationSupport Vector Machines
/9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.
More informationPetri Net Based Software Dependability Engineering
Proc. RELECTRONIC 95, Budapest, pp. 181-186; October 1995 Petr Net Based Software Dependablty Engneerng Monka Hener Brandenburg Unversty of Technology Cottbus Computer Scence Insttute Postbox 101344 D-03013
More informationImproved Resource Allocation Algorithms for Practical Image Encoding in a Ubiquitous Computing Environment
JOURNAL OF COMPUTERS, VOL. 4, NO. 9, SEPTEMBER 2009 873 Improved Resource Allocaton Algorthms for Practcal Image Encodng n a Ubqutous Computng Envronment Manxong Dong, Long Zheng, Kaoru Ota, Song Guo School
More informationCommunication-Minimal Partitioning and Data Alignment for Af"ne Nested Loops
Communcaton-Mnmal Parttonng and Data Algnment for Af"ne Nested Loops HYUK-JAE LEE 1 AND JOSÉ A. B. FORTES 2 1 Department of Computer Scence, Lousana Tech Unversty, Ruston, LA 71272, USA 2 School of Electrcal
More informationA Fast Visual Tracking Algorithm Based on Circle Pixels Matching
A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng
More informationScheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research
Schedulng Remote Access to Scentfc Instruments n Cybernfrastructure for Educaton and Research Je Yn 1, Junwe Cao 2,3,*, Yuexuan Wang 4, Lanchen Lu 1,3 and Cheng Wu 1,3 1 Natonal CIMS Engneerng and Research
More informationSome material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier
Some materal adapted from Mohamed Youns, UMBC CMSC 611 Spr 2003 course sldes Some materal adapted from Hennessy & Patterson / 2003 Elsever Scence Performance = 1 Executon tme Speedup = Performance (B)
More informationLearning-Based Top-N Selection Query Evaluation over Relational Databases
Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **
More informationSimple March Tests for PSF Detection in RAM
Smple March Tests for PSF Detecton n RAM Ireneusz Mroze Balysto Techncal Unversty Computer Scence Department Wejsa 45A, 5-35 Balysto POLAND mroze@.pb.balysto.pl Eugena Buslowsa Balysto Techncal Unversty
More informationParallel and Distributed Association Rule Mining - Dr. Giuseppe Di Fatta. San Vigilio,
Parallel and Dstrbuted Assocaton Rule Mnng - Dr. Guseppe D Fatta fatta@nf.un-konstanz.de San Vglo, 18-09-2004 1 Overvew Assocaton Rule Mnng (ARM) Apror algorthm Hgh Performance Parallel and Dstrbuted Computng
More informationAgenda & Reading. Simple If. Decision-Making Statements. COMPSCI 280 S1C Applications Programming. Programming Fundamentals
Agenda & Readng COMPSCI 8 SC Applcatons Programmng Programmng Fundamentals Control Flow Agenda: Decsonmakng statements: Smple If, Ifelse, nested felse, Select Case s Whle, DoWhle/Untl, For, For Each, Nested
More informationLoad Balancing for Hex-Cell Interconnection Network
Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,
More informationy and the total sum of
Lnear regresson Testng for non-lnearty In analytcal chemstry, lnear regresson s commonly used n the constructon of calbraton functons requred for analytcal technques such as gas chromatography, atomc absorpton
More informationOn Embedding and NP-Complete Problems of Equitable Labelings
IOSR Journal o Mathematcs (IOSR-JM) e-issn: 78-578, p-issn: 39-765X Volume, Issue Ver III (Jan - Feb 5), PP 8-85 wwwosrjournalsorg Ombeddng and NP-Complete Problems o Equtable Labelngs S K Vadya, C M Barasara
More informationThe Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique
//00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy
More informationEnhanced AMBTC for Image Compression using Block Classification and Interpolation
Internatonal Journal of Computer Applcatons (0975 8887) Volume 5 No.0, August 0 Enhanced AMBTC for Image Compresson usng Block Classfcaton and Interpolaton S. Vmala Dept. of Comp. Scence Mother Teresa
More informationAn Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation
17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed
More informationSURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB
SURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB V. Hotař, A. Hotař Techncal Unversty of Lberec, Department of Glass Producng Machnes and Robotcs, Department of Materal
More informationand NSF Engineering Research Center Abstract Generalized speedup is dened as parallel speed over sequential speed. In this paper
Shared Vrtual Memory and Generalzed Speedup Xan-He Sun Janpng Zhu ICASE NSF Engneerng Research Center Mal Stop 132C Dept. of Math. and Stat. NASA Langley Research Center Msssspp State Unversty Hampton,
More informationDiscrete and Continuous Time High-Order Markov Models for Software Reliability Assessment
Dscrete and Contnuous Tme Hgh-Order Markov Models for Software Relablty Assessment Vtaly Yakovyna and Oksana Nytrebych Software Department, Lvv Polytechnc Natonal Unversty, Lvv, Ukrane vtaly.s.yakovyna@lpnu.ua,
More informationEmpirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap
Int. Journal of Math. Analyss, Vol. 8, 4, no. 5, 7-7 HIKARI Ltd, www.m-hkar.com http://dx.do.org/.988/jma.4.494 Emprcal Dstrbutons of Parameter Estmates n Bnary Logstc Regresson Usng Bootstrap Anwar Ftranto*
More information2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements
Module 3: Element Propertes Lecture : Lagrange and Serendpty Elements 5 In last lecture note, the nterpolaton functons are derved on the bass of assumed polynomal from Pascal s trangle for the fled varable.
More informationMeta-heuristics for Multidimensional Knapsack Problems
2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,
More informationReliability and Performance Models for Grid Computing
Relablty and Performance Models for Grd Computng Yuan-Shun Da,2, Jack Dongarra,3,4 Department of Electrcal Engneerng and Computer Scence, Unversty of Tennessee, Knoxvlle 2 Department of Industral and Informaton
More informationFuzzy Modeling of the Complexity vs. Accuracy Trade-off in a Sequential Two-Stage Multi-Classifier System
Fuzzy Modelng of the Complexty vs. Accuracy Trade-off n a Sequental Two-Stage Mult-Classfer System MARK LAST 1 Department of Informaton Systems Engneerng Ben-Guron Unversty of the Negev Beer-Sheva 84105
More informationSequential search. Building Java Programs Chapter 13. Sequential search. Sequential search
Sequental search Buldng Java Programs Chapter 13 Searchng and Sortng sequental search: Locates a target value n an array/lst by examnng each element from start to fnsh. How many elements wll t need to
More informationA New Approach For the Ranking of Fuzzy Sets With Different Heights
New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays
More informationUSING GRAPHING SKILLS
Name: BOLOGY: Date: _ Class: USNG GRAPHNG SKLLS NTRODUCTON: Recorded data can be plotted on a graph. A graph s a pctoral representaton of nformaton recorded n a data table. t s used to show a relatonshp
More informationShared Virtual Memory Machines. Mississippi State, MS Abstract
Performance Consderatons of Shared Vrtual Memory Machnes Xan-He Sun Janpng Zhu Department of Computer Scence NSF Engneerng Research Center Lousana State Unversty Dept. of Math. and Stat. Baton Rouge, LA
More informationChapter 1. Introduction
Chapter 1 Introducton 1.1 Parallel Processng There s a contnual demand for greater computatonal speed from a computer system than s currently possble (.e. sequental systems). Areas need great computatonal
More informationF Geometric Mean Graphs
Avalable at http://pvamu.edu/aam Appl. Appl. Math. ISSN: 1932-9466 Vol. 10, Issue 2 (December 2015), pp. 937-952 Applcatons and Appled Mathematcs: An Internatonal Journal (AAM) F Geometrc Mean Graphs A.
More informationOptimal connection strategies in one- and two-dimensional associative memory models
Optmal connecton strateges n one- and two-dmensonal assocatve memory models Lee Calcraft, Rod Adams, and Nel Davey School of Computer Scence, Unversty of Hertfordshre College lane, Hatfeld, Hertfordshre
More informationMotivation. EE 457 Unit 4. Throughput vs. Latency. Performance Depends on View Point?! Computer System Performance. An individual user wants to:
4.1 4.2 Motvaton EE 457 Unt 4 Computer System Performance An ndvdual user wants to: Mnmze sngle program executon tme A datacenter owner wants to: Maxmze number of Mnmze ( ) http://e-tellgentnternetmarketng.com/webste/frustrated-computer-user-2/
More informationA Genetic Algorithm Based Dynamic Load Balancing Scheme for Heterogeneous Distributed Systems
Proceedngs of the Internatonal Conference on Parallel and Dstrbuted Processng Technques and Applcatons, PDPTA 2008, Las Vegas, Nevada, USA, July 14-17, 2008, 2 Volumes. CSREA Press 2008, ISBN 1-60132-084-1
More informationEfficient Content Distribution in Wireless P2P Networks
Effcent Content Dstrbuton n Wreless P2P Networs Qong Sun, Vctor O. K. L, and Ka-Cheong Leung Department of Electrcal and Electronc Engneerng The Unversty of Hong Kong Pofulam Road, Hong Kong, Chna {oansun,
More informationSteps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices
Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between
More informationAnalysis on the Workspace of Six-degrees-of-freedom Industrial Robot Based on AutoCAD
Analyss on the Workspace of Sx-degrees-of-freedom Industral Robot Based on AutoCAD Jn-quan L 1, Ru Zhang 1,a, Fang Cu 1, Q Guan 1 and Yang Zhang 1 1 School of Automaton, Bejng Unversty of Posts and Telecommuncatons,
More informationAir Transport Demand. Ta-Hui Yang Associate Professor Department of Logistics Management National Kaohsiung First Univ. of Sci. & Tech.
Ar Transport Demand Ta-Hu Yang Assocate Professor Department of Logstcs Management Natonal Kaohsung Frst Unv. of Sc. & Tech. 1 Ar Transport Demand Demand for ar transport between two ctes or two regons
More informationAn Image Fusion Approach Based on Segmentation Region
Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua
More informationPrivate Information Retrieval (PIR)
2 Levente Buttyán Problem formulaton Alce wants to obtan nformaton from a database, but she does not want the database to learn whch nformaton she wanted e.g., Alce s an nvestor queryng a stock-market
More information