Dependability and performance measures for the database practitioner

Teorey, Toby J.; Ng, Wee Teck

doi:10.1109/69.687981

Cited by 8 publications

(3 citation statements)

References 7 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the reliability of transaction in the on‐demand computing system considering the conditional steady‐state user‐perceived availability of resources, is computed as

R_{k, k b, D M_{i}, A_{λ}} false(X false) = ([], \prod_{k = 1}^{n} R_{k} false(X false) . A_{λ}) . ([], \prod_{k = 1}^{n - 1} \prod_{b > k} R_{k b} false(X false)) . ([], \prod_{i = 1}^{m} R_{D M_{i}} false(X false))

where A λ is the steady‐state availability of the resources under the load λ . We take its formula from Mainkar, which is expressed as

A_{λ} = \sum_{c = 1}^{n} A_{c, λ} Q_{c}

where ∀ c =1,…, n and A c , λ of available servers, which has been computed as

\sum_{i = 0}^{K} r_{i} Π_{i}

with the steady‐state probabilities for the model, which are given as

Π_{i} = \frac{{false(c ρ false)}^{i}}{i!} Π_{0}, 1 ⩽ i ⩽ c - 1

Π_{i} = \frac{c^{c} ρ^{i}}{c!} Π_{0}, i ⩾ c

In the above formulation,

ρ = \frac{λ}{c μ}

and Q c have been expressed as

Q_{c} = \frac{n!}{c! false(n - c false)!}

…”

Section: Problem Formulationmentioning

confidence: 99%

See 1 more Smart Citation

Balanced task allocation in the on‐demand computing‐based transaction processing system using social spider optimization

Mahato

Singh

2017

Concurrency and Computation

View full text Add to dashboard Cite

Summary Balanced task allocation is one of the methods that can be used to maximize the performance and reliability in the on‐demand computing‐based transaction processing system. On‐demand computing is an increasingly popular enterprise model. It provides computing resources to the user as needed, which may be maintained within the user's enterprise, or made available by a service provider. The balanced task allocation in such environment is known to be an NP hard. The reliability is a measure of trustworthiness of the system while executing the task. So we derive the reliability formula for the on‐demand computing‐based transaction processing system considering resource availability. We propose the balanced task allocation based on social spider optimization (LBTA_SSO) method for this problem. The LBTA_SSO is based on the cooperative behavior of social‐spiders to find a collection of task allocation solutions. We modified five existing algorithms to obtain the task allocation algorithms; Honey Bee Optimization (HBO), Ant Colony Optimization (ACO), Hierarchical Load Balanced Algorithm (HLBA), Dynamic and Decentralized Load Balancing (DLB), and Randomized Algorithm respectively. Then, we compared the proposed algorithm with these modified algorithms. The results show that our algorithm works better than the modified existing algorithms.

show abstract

“…Finally, the reliability of transaction in the on‐demand computing system considering the conditional steady‐state user‐perceived availability of resources, is computed as

R_{k, k b, D M_{i}, A_{λ}} false(X false) = ([], \prod_{k = 1}^{n} R_{k} false(X false) . A_{λ}) . ([], \prod_{k = 1}^{n - 1} \prod_{b > k} R_{k b} false(X false)) . ([], \prod_{i = 1}^{m} R_{D M_{i}} false(X false))

where A λ is the steady‐state availability of the resources under the load λ . We take its formula from Mainkar, which is expressed as

A_{λ} = \sum_{c = 1}^{n} A_{c, λ} Q_{c}

where ∀ c =1,…, n and A c , λ of available servers, which has been computed as

\sum_{i = 0}^{K} r_{i} Π_{i}

with the steady‐state probabilities for the model, which are given as

Π_{i} = \frac{{false(c ρ false)}^{i}}{i!} Π_{0}, 1 ⩽ i ⩽ c - 1

Π_{i} = \frac{c^{c} ρ^{i}}{c!} Π_{0}, i ⩾ c

In the above formulation,

ρ = \frac{λ}{c μ}

and Q c have been expressed as

Q_{c} = \frac{n!}{c! false(n - c false)!}

…”

Section: Problem Formulationmentioning

confidence: 99%

“…Finally, the reliability of transaction in the on-demand computing system considering the conditional steady-state user-perceived availability of resources 40 , is computed as…”

Section: Reliability Modelmentioning

confidence: 99%

Balanced task allocation in the on‐demand computing‐based transaction processing system using social spider optimization

Mahato

Singh

2017

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…The parameters inserted in the CTMC models (Figures and ) are as follows: failure parameters λ _ O , λ _ B , and λ _ J have equal values and were obtained from Dantas et al ; as the repair parameters μ _ O , μ _ B , and μ _ J found in Dantas et al , these values were chosen because they are values that correspond to fault and repair of systems similar to the one assessed in this paper. Parameters μ _ D and λ _ D were adopted from Teorey and Ng, who perform a dependability and performance analysis in centralized or distributed database. Failover f μ _ J repair rate was defined as an approximate value for automatic recovery, and F parameter, the probability of successful failover, was taken from Bauer et al that uses this value to probability of success of an automatic recovery process.…”

Section: Experimental Studymentioning

confidence: 99%

Availability Evaluation and Sensitivity Analysis of a Mobile Backend‐as‐a‐service Platform

Costa

Araújo

Dantas

et al. 2015

Quality & Reliability Eng

View full text Add to dashboard Cite

Performance evaluation of mobile applications has received considerable attention as a prominent activity for improving services quality. Because many data stored on mobile device are synchronized with distributed data centers, the system availability is a critical attribute that requires investigation. Mobile backend‐as‐a‐service (MBaaS) allows developers to link the backend of their applications to cloud storage, as well as providing device management and integration with social networking services. The OpenMobster platform offers a complete synchronization service for mobile applications, but its availability is an inherent critical issue, because one failure can result in losses for companies that use this environment. Analytical models can be used to assess availability of this type of environment and perhaps mitigate downtimes. This paper proposes a hierarchical model to assess the availability of the MBaaS OpenMobster platform focusing on two scenarios: the basic architecture and the automatic recovery process. The designed models were validated through testbed measurements by automatically injecting and repairing the infrastructure. Taking into account the three layers: hardware, operating system, and the MBaaS OpenMobster, we observed OpenMobster being the most critical service component. We have applied failover strategy on the Java virtual machine, and we obtained 10% of reduction in annual downtime. This work may guide systems' administrators in planning their maintenance policies. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Load balanced scheduling and reliability modeling of grid transaction processing system using colored Petri nets

Mahato¹,

Singh

2019

ISA Transactions

View full text Add to dashboard Cite

Dependability and performance measures for the database practitioner

Cited by 8 publications

References 7 publications

Balanced task allocation in the on‐demand computing‐based transaction processing system using social spider optimization

Balanced task allocation in the on‐demand computing‐based transaction processing system using social spider optimization

Availability Evaluation and Sensitivity Analysis of a Mobile Backend‐as‐a‐service Platform

Load balanced scheduling and reliability modeling of grid transaction processing system using colored Petri nets

Contact Info

Product

Resources

About