Performance and Scalability assessment
This page shows the work in progress with regards to the assessment of performance and scalability of the LarKC Platform.
So far the Data Layer has been validated with regards to its performance and scalability. Results are documented in LarKC deliverables D5.5.1 and D5.5.2
A definition of scalability within LarKC and a first analysis of scalability dimensions is presented in deliverable D5.3.2
Urban Computing use case evaluation is documented in deliverable http://www.larkc.eu/wp-content/uploads/2008/01/larkc_d66-2nd-periodic-report-on-data-and-performances_final.pdf. Some of the measurements presented here are independent from the use case and depend on the current platform implementation. This will be used as input for the platform evaluation.
D1.4.1 Initial framework for measuring and evaluating heuristic problem solving
Scalability dimensions, as presented in D5.3.2, as a basis for discussion
ID |
Scalability Dimension (to be improved) |
Fixed Dimension(s) |
Metric |
Variable Dimension(s) |
Strategy to achieve improvement |
Boundaries |
S1 |
Size of the problem |
Complexity of the problem |
Workflow execution time |
Number/nature of supported hw resources |
Distribution of plug-ins to more powerful resources |
Workflow execution time ≤ (Problem_Size_Final / Problem_ Size_Initial) |
Concurrency of user requests |
Number of supported plug-ins |
Parallelization “between plug-ins” |
||||
Volume of total data stored |
|
Parallelization “within a plug-in” |
||||
|
|
Data Layer partitioning |
||||
S2 |
Complexity of the problem |
Size of the problem (volume of data to be processed per user request) |
Workflow execution time |
Number/nature of supported hw resources |
Distribution of plug-ins to more powerful resources |
|
Concurrency of user requests |
Number of supported plug-ins |
Parallelization “between plug-ins” |
||||
Volume of total data stored |
|
Parallelization “within a plug-in” |
||||
|
|
Data Layer partitioning |
||||
S3 |
Concurrency of user requests |
Complexity of the problem |
Workflow execution time |
Number/nature of supported hw resources |
Data Layer redundancy and load balancing |
Workflow execution time ≤ (Number_Users_Final / Number_Users_Initial) |
Size of the problem (volume of data to be processed per user request) |
|
|
||||
Volume of total data stored |
|
|
||||
Number of supported plug-ins |
|
|
||||
S4 |
Volume of total data indexed/processed/stored |
Complexity of the problem |
Setup / indexing / preprocessing time |
Number/nature of supported hw resources |
Data Layer redundancy and load balancing |
|
Size of the problem (volume of data to be processed per user request) |
||||||
Concurrency of user requests |
||||||
Volume of total data stored |
||||||
Number of supported plug-ins |
||||||
S5 |
Number/nature of supported hw resources |
Complexity of the problem |
Number/nature of supported hw resources |
Platform configuration: support of resources through platform adapters |
||
Size of the problem (volume of data to be processed per user request) |
||||||
Concurrency of user requests |
||||||
Volume of total data stored |
||||||
Number of supported plug-ins |
||||||
S6 |
Number of supported plug-ins |
Complexity of the problem |
||||
Size of the problem (volume of data to be processed per user request) |
||||||
Concurrency of user requests |
||||||
Volume of total data stored |