Quality of service enabled database applications
read more
Citations
How a consumer can measure elasticity for cloud platforms
PQR: Predicting Query Execution Times for Autonomous Workload Management
Adaptive quality of service management for enterprise services
Dynamic workload management for very large data warehouses: juggling feathers and bowling balls
Quality of Service-enabled Management of Database Workloads.
References
Transaction Processing: Concepts and Techniques
Toward autonomic web services trust and selection
A method for transparent admission control and request scheduling in e-commerce web sites
Querying business processes with BP-QL
How to Determine a Good Multi-Programming Level for External Scheduling
Related Papers (5)
Customer-defined service level agreements for composite applications
Frequently Asked Questions (13)
Q2. What is the purpose of this paper?
The contribution of this paper is to enable QoS for the bottom layer of a service infrastructure, where almost all services access a shared database.
Q3. What is the reason why the SLA conformances are delayed?
as requests stemming from low-priority terminals do not have deadlines, these requests are delayed as long as possible to allow the prioritized execution of higher priority requests.
Q4. How is the time constraint computed for a request?
The time constraint enfi for the current request is computed by subtracting the observed execution times and the expected time to process the remaining requests from xd.
Q5. Why is the workload of the database based on the multitude of services?
Due to the multitude of services which access the database, the workload of the database consists of requests stemming from many different customers with different service classes, each having a dedicated SLA.
Q6. What is the purpose of the TPC-C benchmark?
The TPC-C-benchmark models a company which is a wholesale supplier operating several warehouses which serve customers in geographically distributed sales districts.
Q7. What is the importance of the SLA?
Scheduling is based on adaptive priorities which are derived from the current level of conformance with the request’s SLA, that is, the percentage of timely requests, and the economic importance of this SLA relative to other pending requests’
Q8. What are the advantages of squared terms?
These studies have shown that squared terms were better suited to model the opportunity costs and marginal gains than linear order higher order terms.
Q9. What is the penalty of a request that is delayed?
If all of these requests are delayed, e.g., by waiting for database locks, the SLA conformance falls onto the next lower service level.
Q10. How many terminals are there in the TPC-C benchmark?
As specified by the TPC-C, the number of terminals is ten times the number of warehouses, thus yielding a total number of 200 terminals during the benchmark.
Q11. How did the authors test the effectiveness of their proposed approach?
Using their prototype, the authors demonstrated the effectiveness of their proposed approach by performing comprehensive real-world studies using the TPC-C benchmark as OLTP workload.
Q12. What is the rationale for choosing squared terms?
The marginal gain mg is a piece-wise quadratic function:mg(c) := (c−cn+1cn−cn+1)2· ∆n−1, cn+1 ≤ c < cn· · · (c−c3 c2−c3)2· ∆1, c3 ≤ c < c20, otherwiseAnalogous to the opportunity costs, the rationale for choosing squared terms is given below.
Q13. What is the SLA conformance for the TPC-C benchmark?
the authors present the analysis of the SLA conformance using static prioritization, that is, the priority of a customer remains constant throughout the entire benchmark.