【摘要】介紹了一種區(qū)分任務(wù)類型的負(fù)載平衡算法。在該算法中，按照占用的系統(tǒng)資源狀況，任務(wù)被劃分成若干類型。調(diào)度時(shí)，各節(jié)點(diǎn)的負(fù)載依照待調(diào)度任務(wù)的任務(wù)類型和當(dāng)前系統(tǒng)資源的負(fù)載動(dòng)態(tài)計(jì)算，并依照計(jì)算結(jié)果在系統(tǒng)的一個(gè)子集范圍內(nèi)尋找適合解。該算法可以避免因?yàn)?a target='_blank' class='arckwlink_none'>信息延遲造成的群聚效應(yīng)，并可以滿足少量對響應(yīng)時(shí)間有特殊規(guī)定的任務(wù)的需求,已經(jīng)用于自主設(shè)計(jì)的分布式操作系統(tǒng)DPOS，效果良好。
關(guān) 鍵 詞 負(fù)載平衡; 區(qū)分任務(wù)類型; 分布式系統(tǒng); 調(diào)度; 群聚效應(yīng)波

We present a novel load-balancing algorithm to the distributed system. In the algorithm, tasks are classified by the resource which they would take up later. While scheduling, every node’s load is calculated dynamically through the task type and the resource’s overhead collected periodically. Then a suitable node is found to the task in a subset of the whole nodes. The algorithm can remove the herd effect caused by delayed information, and can decrease the mean response time of a few tasks that have special requirement in this regard. It has been adopted in a distributed operation system developed by our group.