Algorithms for Dispersed Processing Josef Spillner, Alexander Schill - - PowerPoint PPT Presentation

▶
algorithms for dispersed processing
SMART_READER_LITE
LIVE PREVIEW

Algorithms for Dispersed Processing Josef Spillner, Alexander Schill - - PowerPoint PPT Presentation

Jan 19, 2024 279 likes •444 views

Department of Computer Science | Institute of Systems Architecture | Chair of Computer Networks Algorithms for Dispersed Processing Josef Spillner, Alexander Schill mailto:josef.spillner@tu-dresden.de xmpp:josef.spillner@jabber.org 1 st

slide-1
SLIDE 1

Department of Computer Science | Institute of Systems Architecture | Chair of Computer Networks

Algorithms for Dispersed Processing

1st International Workshop on Advances in Cloud Computing Legislation, Accountability, Security and Privacy (CLASP), December 8-11, 2014, London, UK

Josef Spillner, Alexander Schill mailto:josef.spillner@tu-dresden.de

xmpp:josef.spillner@jabber.org

slide-2
SLIDE 2

# 2 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Background: Informatjon Dispersal

Motivation »Don't trust the cloud.« Risks:

  • temporary unavailability
  • permanent unavailability
  • of data [loss]
  • of service [bankruptcy]
  • arbitrary slowness
  • unauthorised access [honest-but-curious] || 3rd-party spying
  • malicious modification
  • refusal to delete
  • lock-in, transfer costs
  • ... etc. pp.

Insufficient but essential protection: Information dispersal over multiple clouds. => IDA

slide-3
SLIDE 3

# 3 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Background: Informatjon Dispersal

Definition Popular definition: But: Why just sit or traverse? Storage « √ Networking « √ Processing « ???

[Image source: techtarget.com]

slide-4
SLIDE 4

# 4 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Background: Informatjon Dispersal

Generalised Dispersed Processing Storage Network (Communication) Processing (Computation) data locality

slide-5
SLIDE 5

# 5 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Map-Reduce

(a), (c): central processing (b), (d): map-reduce IDA of choice: bitsplitting; e.g. 50% redundancy: k=2, m=1, n=k+m Property: structure-preserving

slide-6
SLIDE 6

# 6 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Summatjon Algorithm

Addition: requires n=2 Multiplication: requires n=4 (unless a, b ≠ 0) Method:

  • map: sum/mult
  • reduce: sum
slide-7
SLIDE 7

# 7 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Search Algorithm

Pattern search without index Method:

  • split pattern, too
  • map: perform k partial searches
  • reduce: filter out false positives
slide-8
SLIDE 8

# 8 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Map-Carry-Reduce

Limits of parallelisation

  • fixed-size integer shifts

→ carry bits

  • filter operations

→ position data Security effect: Information leakage

slide-9
SLIDE 9

# 9 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Encryptjon

Processing dispersed + encrypted data blocks Property: structure-preserving! Algorithms:

  • homomorphic encryption

→ arithmetics

  • order-preserving encryption

→ sorting

  • convergent encryption

→ deduplication Quelle Speicher Quelle Speicher Verschlüsselung ⚷ Source Cloud Encryption ⚷ Splitting Cloud Ѧ Policies

slide-10
SLIDE 10

# 10 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Encryptjon Example

Travel distance: 71 km + 19 km

  • = 90 km

71 km 19 km 283060154 630596813 Travel distance: 540987952 „km“ = 90 km kpub kpub Arithmetic: a*b%bits 283060154 “km“ + 630596813 “km“

  • = 540987952 “km“

generate-keypairbits=16 => {kpriv, kpub} kpub, kpriv

slide-11
SLIDE 11

# 11 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Precision

If a cloud fails... ... repair and continue processing? ... or, expect degraded results? Application-specific decision. Example: floating point with (i)nteger, (f)ractional, (r)edundant parts.

slide-12
SLIDE 12

# 12 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Redundancy

Redundant data is not generally processable. Distribution matters.

slide-13
SLIDE 13

# 13 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Dispersed Processing: Algorithms Overview

Classification of algorithms (pending)

slide-14
SLIDE 14

# 14 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Performance Evaluatjon

slide-15
SLIDE 15

# 15 1st CLASP , 11.12.2014 Algorithms for Dispersed Processing

  • J. Spillner & A. Schill

Conclusion & Stealth Roadmap

Towards Dispersed Cloud Computing [BSC'14] Towards Dispersed Cloud Computing [CLASP'14] Stealth Apps for Secure Personal Data Analytics [NetSys'15] Stealth Databases ...

Dispersed Computing Storage, networking: Much existing research Processing: need for special algorithms Evaluation: slower processing vs. (often) less transmission Code: git://nubisave.org/git/dispersedalgorithms Stealth Computing Combination of dispersion x encryption and further quality measures Enabler for native cloud applications