For Project Enquiry +91 9962 067 067

EXISTING SYSTEM

PROPOSED SYSTEM

EXISTING CONCEPT:-

In the existing system guaranteed compute resources always ready to accommodate job execution under reserved instance pricing.

On-demand compute resources, which may delay launching job execution until they are available but on-going execution, will not be aborted, under on-demand pricing.

Bidden compute resources, which abort execution upon price hikes beyond the bidden amount, under spot instance pricing.

PROPOSED CONCEPT:-

In the proposed system aims (SI) pricing are least expensive, albeit to reduced availability. While SI pricing gives clients with chances to rent more compute resources, check pointing is highly desired for such compute resources to boost overall execution performance and productivity, due to their relatively frequent resource unavailability resulting from revocation, as explored lately.

Check pointing saves the states of program execution to let it resume execution from the last checkpoint upon hardware failures or SI revocation

EXISTING TECHNIQUE:-

Adjusted Markov Model (AMM)

PROPOSED TECHNIQUE:-

Enhanced adaptive incremental check pointing (EAIC)

TECHNIQUE DEFNITION:-

Cloud providers usually offer their customers options to purchase instances with variable levels of availability and prices.

TECHNIQUE DEFNITION:-

EAIC model handles SI revocation and three failure levels with two checkpoint types, where is the time to do actual work, is the checkpoint time of failure level, is the recovery time from, and is the average time since SI revocation until the price falls below the user bid again.

DRAWBACKS:-

AMM to uncover potential benefits of

Checkpointing in RaaS clouds.

ADVANTAGES:-

Fast and accurately predicting the desirable points of time to take checkpoints so as to markedly reduce the expected job turnaround time and the monetary cost.