#ATAGTR2021 Presentation : "Performance Evaluation Strategy of multi-access edge computing" by Arnab Majumdar

Agile Testing Alliance
Agile Testing AllianceAgile Testing Alliance
#ATAGTR2021
Performance Evaluation
Strategy of Edge Computing
Author code – 047
Arnab Majumdar
Performance Evaluation Strategy of multi-access edge computing
• Introduction
• MEC- Advantages and Challenges
• Performance Model
• Deployment Model
• Testing Approach and Environment
• Conclusion
• MEC – Use
Introduction – multi access edge computing
• Distributed Computing system
• Brings data storage and computation
closer to data sources
• Low latency in network response
• Proportionate redistribution of computing
capability, data, and storage for micro
services
Advantages Security
Speed
Scalability
Versatility
Cost Effective
Reliability
MEC - challenges
Edge computing clouds are intrinsically local and have a smaller scale and are thus
subject to significantly larger fluctuations in offered traffic due to factors such as
correlated events and user mobility, thus making the network traffic throughput
unpredictable.
Edge computing systems by definition are distributed across multiple edge networks and
hence are associated with considerable heterogeneity in bandwidth and compute
resources.
The unique nature of the distributed edge cloud system poses key design challenges
such as specification of a control plane for distributed edge, distributed or centralized
resource assignment strategies, traffic load balancing, orchestration of computing
functions and related network routing of data, mobility management technique.
Performance Model
• MECCA – Mobile Edge Cloud Cornering
application
• Recommendation Service – Responsible
for recommending suitable speed for all
upcoming curves around a given location
• External Database - used to permanently
store curve results
• Detection Service - requested to detect
and calculate detailed information about
upcoming curves
• Service Registry - needed to support
services discovery for various service
instances as we have many instances from
the above-mentioned services
Deployment Model
Cloud-deployment model:
all services are deployed in clouds,
whereas clients are in mobile devices. We
distinguish two cases: the application
owner deploys his/her own services (i) not
in the same data center with and (ii) in the
same data center with other third party
services.
Edge-cloud deployment:
Application services are deployed
in both the edge and cloud.
Similarly, application services
might be co-located or not with
third party services.
Cloud-deployment model testing is very
useful for the design of edge-cloud
deployment because we can detect suitable
services in clouds that can be migrated to
the edge or optimized for cloud deployment.
That is the reason why in this paper we also
focus on testing cloud-deployment.
Test Approach and Environment
The core assumption behind
MEC testing strategy is steady
increase of throughput with
more clients in network.
• Response-time
• Result-time
• Response-ratio
• Result-ratio
Key Performance Metrics
Conclusion
Scalability
Transactions
throughput
Transactions
Cost
Responsivenes
s
Use Cases for Edge Computing
• Autonomous vehicles
• Stronger security
• Healthcare
• Manufacturing and industrial processes
• Augmented reality
• Enhanced workplace safety
• Streaming services
#ATAGTR2021 Presentation : "Performance Evaluation Strategy of multi-access edge computing" by Arnab Majumdar
1 von 11

Más contenido relacionado

Similar a #ATAGTR2021 Presentation : "Performance Evaluation Strategy of multi-access edge computing" by Arnab Majumdar

Cloud computingCloud computing
Cloud computingHamza Sajjad
146 views25 Folien
cloudintro-lec01.pptcloudintro-lec01.ppt
cloudintro-lec01.pptTomMot10
8 views20 Folien

Similar a #ATAGTR2021 Presentation : "Performance Evaluation Strategy of multi-access edge computing" by Arnab Majumdar(20)

Cloud computingCloud computing
Cloud computing
Hamza Sajjad146 views
Cloud computing managingCloud computing managing
Cloud computing managing
Universiti Putra Malaysia12 views
Cloud Migration - CCS Technologies (P) Ltd.Cloud Migration - CCS Technologies (P) Ltd.
Cloud Migration - CCS Technologies (P) Ltd.
CCS Technologies (P) Ltd.213 views
cloudintro-lec01.pptcloudintro-lec01.ppt
cloudintro-lec01.ppt
NishantAnand395 views
cloudintro-lec01.pptcloudintro-lec01.ppt
cloudintro-lec01.ppt
TomMot108 views
Migration into a CloudMigration into a Cloud
Migration into a Cloud
Divya S616 views
TERM PAPER presentation (2).pptxTERM PAPER presentation (2).pptx
TERM PAPER presentation (2).pptx
KalashShandilya14 views
Chapter1-IntroductionChapter1-Introduction
Chapter1-Introduction
Namrata Vardhaman71 views
Odbc and data access objectsOdbc and data access objects
Odbc and data access objects
Sangeetha Sg2.1K views
Cloud TestingCloud Testing
Cloud Testing
Binnate E Hawwa2.7K views
Cloud Computing Basics.pptxCloud Computing Basics.pptx
Cloud Computing Basics.pptx
SeethaDinesh11 views

Más de Agile Testing Alliance(20)

Parallel Test execution in Cypress with CI/CDParallel Test execution in Cypress with CI/CD
Parallel Test execution in Cypress with CI/CD
Agile Testing Alliance6 views
Localisation Testing using UI automationLocalisation Testing using UI automation
Localisation Testing using UI automation
Agile Testing Alliance8 views
AI in UI automation using HeleniumAI in UI automation using Helenium
AI in UI automation using Helenium
Agile Testing Alliance10 views
Automation for test data anonymizationAutomation for test data anonymization
Automation for test data anonymization
Agile Testing Alliance6 views
MobSF: Mobile Security Testing (Android/IoS)MobSF: Mobile Security Testing (Android/IoS)
MobSF: Mobile Security Testing (Android/IoS)
Agile Testing Alliance34 views
Web Application Security TestingWeb Application Security Testing
Web Application Security Testing
Agile Testing Alliance13 views
Significance of AI in TestingSignificance of AI in Testing
Significance of AI in Testing
Agile Testing Alliance14 views
Accessibility Testing.pptxAccessibility Testing.pptx
Accessibility Testing.pptx
Agile Testing Alliance12 views

Último(20)

Java Platform Approach 1.0 - Picnic MeetupJava Platform Approach 1.0 - Picnic Meetup
Java Platform Approach 1.0 - Picnic Meetup
Rick Ossendrijver23 views
Green Leaf Consulting: Capabilities DeckGreen Leaf Consulting: Capabilities Deck
Green Leaf Consulting: Capabilities Deck
GreenLeafConsulting170 views
ThroughputThroughput
Throughput
Moisés Armani Ramírez28 views

#ATAGTR2021 Presentation : "Performance Evaluation Strategy of multi-access edge computing" by Arnab Majumdar

  • 1. #ATAGTR2021 Performance Evaluation Strategy of Edge Computing Author code – 047 Arnab Majumdar
  • 2. Performance Evaluation Strategy of multi-access edge computing • Introduction • MEC- Advantages and Challenges • Performance Model • Deployment Model • Testing Approach and Environment • Conclusion • MEC – Use
  • 3. Introduction – multi access edge computing • Distributed Computing system • Brings data storage and computation closer to data sources • Low latency in network response • Proportionate redistribution of computing capability, data, and storage for micro services
  • 5. MEC - challenges Edge computing clouds are intrinsically local and have a smaller scale and are thus subject to significantly larger fluctuations in offered traffic due to factors such as correlated events and user mobility, thus making the network traffic throughput unpredictable. Edge computing systems by definition are distributed across multiple edge networks and hence are associated with considerable heterogeneity in bandwidth and compute resources. The unique nature of the distributed edge cloud system poses key design challenges such as specification of a control plane for distributed edge, distributed or centralized resource assignment strategies, traffic load balancing, orchestration of computing functions and related network routing of data, mobility management technique.
  • 6. Performance Model • MECCA – Mobile Edge Cloud Cornering application • Recommendation Service – Responsible for recommending suitable speed for all upcoming curves around a given location • External Database - used to permanently store curve results • Detection Service - requested to detect and calculate detailed information about upcoming curves • Service Registry - needed to support services discovery for various service instances as we have many instances from the above-mentioned services
  • 7. Deployment Model Cloud-deployment model: all services are deployed in clouds, whereas clients are in mobile devices. We distinguish two cases: the application owner deploys his/her own services (i) not in the same data center with and (ii) in the same data center with other third party services. Edge-cloud deployment: Application services are deployed in both the edge and cloud. Similarly, application services might be co-located or not with third party services. Cloud-deployment model testing is very useful for the design of edge-cloud deployment because we can detect suitable services in clouds that can be migrated to the edge or optimized for cloud deployment. That is the reason why in this paper we also focus on testing cloud-deployment.
  • 8. Test Approach and Environment The core assumption behind MEC testing strategy is steady increase of throughput with more clients in network. • Response-time • Result-time • Response-ratio • Result-ratio Key Performance Metrics
  • 10. Use Cases for Edge Computing • Autonomous vehicles • Stronger security • Healthcare • Manufacturing and industrial processes • Augmented reality • Enhanced workplace safety • Streaming services