Muutke küpsiste eelistusi

E-raamat: Optimized Cloud Based Scheduling

, ,
Teised raamatud teemal:
  • Formaat - PDF+DRM
  • Hind: 55,56 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
Optimized Cloud Based SchedulingSuurem pilt
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

This book presents an improved design for service provisioning and allocation models that are validated through running genome sequence assembly tasks in a hybrid cloud environment. It proposes approaches for addressing scheduling and performance issues in big data analytics and showcases new algorithms for hybrid cloud scheduling. Scientific sectors such as bioinformatics, astronomy, high-energy physics, and Earth science are generating a tremendous flow of data, commonly known as big data. In the context of growing demand for big data analytics, cloud computing offers an ideal platform for processing big data tasks due to its flexible scalability and adaptability. However, there are numerous problems associated with the current service provisioning and allocation models, such as inefficient scheduling algorithms, overloaded memory overheads, excessive node delays and improper error handling of tasks, all of which need to be addressed to enhance the performance of big data analytics.

1 Introduction
1(10)
1.1 Project Overview
1(2)
1.2 Project Description
3(1)
1.3 Project Objectives
4(1)
1.4 Motivations and Significances of the Project
4(5)
1.5 Outline
9(2)
2 Background
11(24)
2.1 Literature Reviews
11(20)
2.1.1 Scale of Things---Peta and Exascale
11(1)
2.1.2 Current Resource Availability
12(1)
2.1.3 High Performance Computing
13(1)
2.1.4 Cloud Computing
14(5)
2.1.5 Existing HPC Implementation Architecture
19(5)
2.1.6 Significance of HPC Clusters on Hybrid Cloud
24(1)
2.1.7 Hybrid Cloud Future Improvements
25(2)
2.1.8 HPC + Cloud Contrasted with Existing Implementation Architectures
27(1)
2.1.9 Profit-Based Scheduling
28(1)
2.1.10 Preemptable Scheduling
29(1)
2.1.11 Priority-Based Scheduling
30(1)
2.2 Various Aspects and Current Issues
31(4)
2.2.1 Service Provisioning
31(1)
2.2.2 Service Allocation
32(1)
2.2.3 Service Adaptation
32(1)
2.2.4 Service Mapping
33(2)
3 Benchmarking
35(12)
3.1 Cloud Benchmarking Instance Specifications and Assumptions
35(1)
3.2 Classic Benchmark Test Categories
36(1)
3.2.1 Dhrystone Benchmark
36(1)
3.2.2 Unpack Benchmark
36(1)
3.2.3 Livermore Loops
37(1)
3.2.4 Whetstone Benchmark
37(1)
3.3 Disk, USB and LAN Benchmarks
37(1)
3.4 Multithreading Benchmarks
38(4)
3.4.1 Simple Add Tests
39(1)
3.4.2 Whetstone Benchmark
39(1)
3.4.3 MP FLOPS Program
40(1)
3.4.4 MP Memory Speed Tests
40(1)
3.4.5 MP Memory Bus Speed Tests
40(1)
3.4.6 MP Memory Random Access Speed Benchmark
40(2)
3.5 OpenMP Benchmarks for Parallel Processing Performance
42(1)
3.5.1 MemSpeed
42(1)
3.5.2 Original Open MP Benchmark
42(1)
3.6 Memory BusSpeed Benchmark
42(5)
3.6.1 BusSpeed Test
42(1)
3.6.2 Random Memory Benchmark
43(1)
3.6.3 SSE Benchmark
43(4)
4 Computation of Large Datasets
47(18)
4.1 Challenges and Considerations in Computing Large-Scale Data
48(2)
4.1.1 Extreme Parallelism and Heterogeneity
48(1)
4.1.2 Data Transfer Costs
48(1)
4.1.3 Increased Failure Rates
49(1)
4.1.4 Power Requirements at Exascale
49(1)
4.1.5 Memory Requirements at Exascale
50(1)
4.2 Work Done to Enable Computation of Large Datasets
50(13)
4.2.1 OpenStack and Azure Implementation
51(5)
4.2.2 Cloud Interconnectivity
56(2)
4.2.3 HPC + Cloud
58(1)
4.2.4 Simulations and Tools for Large Scale Data Processing
59(4)
4.3 Summary, Discussion and Future Directions
63(2)
4.3.1 What Might the Model Look like?
63(1)
4.3.2 Application Primitives---Key to Performance
64(1)
5 Optimized Online Scheduling Algorithms
65(18)
5.1 Dynamic Task Splitting Allocator (DTSA)
65(2)
5.2 Procedural Parallel Scheduling Heuristic (PPSH)
67(8)
5.2.1 Mapping Phase
68(2)
5.2.2 Shuffling Phase
70(3)
5.2.3 Eliminating Phase
73(2)
5.2.4 Finishing Phase
75(1)
5.3 Scalable Parallel Scheduling Heuristic (SPSH)
75(4)
5.4 Calculation of Time
79(4)
6 Performance Evaluation
83(8)
6.1 Complexity Analysis
83(2)
6.1.1 Algorithm 1: Dynamic Task Splitting Allocator (DTSA)
83(1)
6.1.2 Algorithm 2: PPSH Phase 1---Mapping Phase
83(1)
6.1.3 Algorithm 3: PPSH Phase 2---Shuffling Phase
84(1)
6.1.4 Algorithm 4: PPSH Phase 3---Eliminating Phase
84(1)
6.1.5 Algorithm 5: PPSH Phase 4---Finishing Phase
84(1)
6.1.6 Algorithm 6: Scalable Parallel Scheduling Heuristic (SPSH)
84(1)
6.2 Experimental Results
85(6)
7 Conclusion and Future Works
91(2)
Bibliography 93
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97833197321452e.html