Sentence examples for checkpoint time from inspiring English sources

In short, for the operations that occur sufficiently close to the checkpoint time, there is no guarantee that they get ordered identically at both the sides with respect to the checkpoint time.

This allowed us to say that if the checkpoint time used at the provider is known to the consumer, then the consumer can match the ByteHrs figures of the provider.

To attenuate this: (1) we assume that one base-generic running checkpoint image is accessible to all the clients; (2) the applications start their execution on top of this image once it is locally resumed; and (3) at checkpoint time, we only transmit the differences between the current image and the base image.

We expect that for a checkpoint, the provider will directly measure the storage space actually occupied, whereas, for a given checkpoint time, the consumer will mimic the process by adding (for PUT) and subtracting (for DELETE) to calculate the space, and as we discussed with respect to Figure9, discrepancies are possible.

In the discussion concerning calculation of ByteHrs (illustrated using Figure8), we have implicitly assumed that the execution of a PUT (respectively a DELETE) operation is an atomic event whose time of occurrence is either less or greater than the checkpoint time (i.e., the operation happens either before or after the checkpoint).

We hypothesize that due to its independent transcriptional regulation, BMH1's products are present at higher levels than those of BMH2 at checkpoint time.

Ideally, Amazon's checkpoint times should be made known to consumers to prevent any such errors.

Assuming checkpoint times are known to a consumer, then any discrepancies can be resolved at the consumer side by examining the storage consumption figures of the provider and working out the place of the operations that happened around the checkpoint times.

In Figure9 we explicitly show network and operation execution latencies for a given operation, say PUT; also, i, j, k and l are provider side checkpoint times used for illustration.

Any discrepancies that get introduced unintentionally (e.g., due to non identical checkpoint times) can be resolved by consumers by careful examination of corresponding resource usage data from providers.

Moreover, while checkpointing time was considered in their model, the actual checkpointing time on spinning disks, especially for cloud systems that are not specialized for parallel I/O, can represent much more than 10% of overhead, which is the value expected for very large-scale machines such as APEX and EXASCALE.