Checkpointing scheme based on imprecise computation for real - time systems 基于非精確計(jì)算的實(shí)時(shí)任務(wù)檢查點(diǎn)設(shè)置策略
You can optionally have the application do occasional checkpointing , so that when it is rerun after a cancellation it does not need to do all its work over again 您可以選擇非經(jīng)常性地進(jìn)行檢查點(diǎn)操作,這樣當(dāng)應(yīng)用程序中斷之后重新運(yùn)行時(shí),就不需要再次執(zhí)行所有的操作。
Call these around code segments during which you don ' t want to be suspended or killed by the core client . note : this is done automatically while checkpointing 在一段代碼的前后分別調(diào)用這兩個(gè)函數(shù),可以防止這段代碼的執(zhí)行過(guò)程中被搶占或是被客戶端給關(guān)閉進(jìn)程。注意:這些函數(shù)會(huì)在保存進(jìn)度的時(shí)候自動(dòng)被調(diào)用。
A probalistic model is proposed for optimistic simulation based on the adherence to realistic assumptions and characterization of various overhead costs , namely non - negligible communication delay , state - saving costs , rollback costs , infrequent checkpointing and rollback costs and n processors 本文用概率方法提出一個(gè)樂(lè)觀仿真的性能模型,基于切合實(shí)際的假設(shè),包含了通訊延遲、狀態(tài)保存開(kāi)銷、非頻繁檢核、層疊回滾,模型適用于n個(gè)處理器。
Secondly , based on the ftoum computing model , the thesis presents a new fault tolerance algorithm , oraml , using both checkpointing and module replication , which employs flexible configuration management mechanisms to implement dynamic replication , imports the fault tolerance policies on the client side to make the clients take part in the fault tolerance process actively , and enables the fault tolerance process transparent to the clients completely while separating the replication protocol from the communication protocol 2 、現(xiàn)有的分布式容錯(cuò)算法大多缺乏對(duì)客戶方容錯(cuò)策略的支持,并且很難做到既對(duì)客戶透明,又能將復(fù)制協(xié)議和通信協(xié)議有效分離,本文基于ftoum計(jì)算模型,并采取檢查點(diǎn)設(shè)置和模塊復(fù)制相結(jié)合的方法,提出了一種新的動(dòng)態(tài)容錯(cuò)算法oraml 。 oraml算法采用靈活的配置管理機(jī)制實(shí)現(xiàn)動(dòng)態(tài)復(fù)制;引入客戶方容錯(cuò)策略,使得客戶方可以主動(dòng)參與容錯(cuò);實(shí)現(xiàn)了復(fù)制協(xié)議與可靠通信協(xié)議的分離;并且算法所采用的容錯(cuò)機(jī)制對(duì)用戶完全透明。