Hi all, Offline discussed with Zhu Zhu, Yangze Guo, Yuepeng Pan. We reached consensus on slot.request.max-interval and taskmanager.load-balance.mode. And I have updated the FLIP.
For a detailed introduction to taskmanager.load-balance.mode, please refer to FLIP’s 3.1 Public Interfaces[1]. And the strategy for slot.request.max-intervel has been improved. The latest strategy can be referred from FLIP’s 2.2.2 Waiting mechanism[2]. For comparison of old and new strategies, please refer to RejectedAlternatives[3]. Thanks again to everyone who participated in the discussion. Looking forward to your continued feedback. [1] https://cwiki.apache.org/confluence/display/FLINK/FLIP-370%3A+Support+Balanced+Tasks+Scheduling#FLIP370:SupportBalancedTasksScheduling-3.1PublicInterfaces [2] https://cwiki.apache.org/confluence/display/FLINK/FLIP-370%3A+Support+Balanced+Tasks+Scheduling#FLIP370:SupportBalancedTasksScheduling-2.2.2Waitingmechanism [3] https://cwiki.apache.org/confluence/display/FLINK/FLIP-370%3A+Support+Balanced+Tasks+Scheduling#FLIP370:SupportBalancedTasksScheduling-RejectedAlternatives Best, Rui On Thu, Oct 12, 2023 at 9:49 AM Yuepeng Pan <panyuep...@apache.org> wrote: > Hi, Shammon. > Thanks for your feedback. > > >1. This mechanism will be only supported in `SlotPool` or both `SlotPool` > and `DeclarativeSlotPool`? > > As described on the FLIP page, the current design plans to introduce the > waiting mechanism only in the `SlotPool`, because the existing > `WaitingForResources` can already achieve this effect. > > >Currently the two slot pools are used in different schedulers. > > Yes, that's indeed the case. > > >I think this will also bring value to `DeclarativeSlotPool`, but > currently FLIP content seems to be based on `SlotPool`, right? > > Yes. your expectations are indeed reasonable. In theory, the > `DeclarativeSlotPool` could also benefit from a waiting mechanism, as > discussed. The purpose of introducing the waiting mechanism is to enable > the `SlotPool` to have a global view to calculate the globally optimal > solution. I've rechecked the relevant logic in the `AdaptiveScheduler`, and > as I understand, the existing mechanisms already fulfill the current > feature requirements. You could find more conclusions on this in FLIP > `3.2.5`. Of course, I'd be appreciated with your confirmation. If there's > any misunderstanding on my part, please correct me. > > >2. ... What should be done when the slot selected by the round-robin > strategy cannot meet the resource requirements? > > Is this referring to the phase of task-to-slot allocation? I'm not quite > sure, would you mind explaining it? Thanks~. > > >3. Is the assignment of tasks to slots balanced based on region or job > level? > > Currently, there is no specific handling based on regions, and there is no > job-level balancing. The target effect of the current feature is to achieve > load balancing based on the number of tasks at the Task Manager (TM) level. > Looking forward to any suggestions regarding the item you mentioned. > > >When multiple TMs fail over, will it cause the balancing strategy to fail > or even worse? > > IIUC, when multiple Task Managers undergo failover, the results after > successful recovery will still be maintained in a relatively balanced state. > > >What is the current processing strategy? > > The Slot-to-TM strategy does not change after a Task Manager undergoes > failover. > > Best, Regards. > Yuepeng Pan > > On 2023/09/28 05:10:13 Shammon FY wrote: > > Thanks Yuepeng for initiating this discussion. > > > > +1 in general too, in fact we have implemented a similar mechanism > > internally to ensure a balanced allocation of tasks to slots, it works > well. > > > > Some comments about the mechanism > > > > 1. This mechanism will be only supported in `SlotPool` or both `SlotPool` > > and `DeclarativeSlotPool`? Currently the two slot pools are used in > > different schedulers. I think this will also bring value to > > `DeclarativeSlotPool`, but currently FLIP content seems to be based on > > `SlotPool`, right? > > > > 2. In fine-grained resource management, we can set different resource > > requirements for different nodes, which means that the resources of each > > slot are different. What should be done when the slot selected by the > > round-robin strategy cannot meet the resource requirements? Will this > lead > > to the failure of the balance strategy? > > > > 3. Is the assignment of tasks to slots balanced based on region or job > > level? When multiple TMs fail over, will it cause the balancing strategy > to > > fail or even worse? What is the current processing strategy? > > > > For Zhuzhu and Rui: > > > > IIUC, the overall balance is divided into two parts: slot to TM and task > to > > slot. > > 1. Slot to TM is guaranteed by SlotManager in ResourceManager > > 2. Task to slot is guaranteed by the slot pool in JM > > > > These two are completely independent, what are the benefits of unifying > > these two into one option? Also, do we want to share the same > > option between SlotPool in JM and SlotManager in RM? This sounds a bit > > strange. > > > > Best, > > Shammon FY > > > > > > > > On Thu, Sep 28, 2023 at 12:08 PM Rui Fan <1996fan...@gmail.com> wrote: > > > > > Hi Zhu Zhu, > > > > > > Thanks for your feedback here! > > > > > > You are right, user needs to set 2 options: > > > - cluster.evenly-spread-out-slots=true > > > - slot.sharing-strategy=TASK_BALANCED_PREFERRED > > > > > > Update it to one option is useful at user side, so > > > `taskmanager.load-balance.mode` sounds good to me. > > > I want to check some points and behaviors about this option: > > > > > > 1. The default value is None, right? > > > 2. When it's set to Tasks, how to assign slots to TM? > > > - Option1: It's just check task number > > > - Option2: It''s check the slot number first, then check the > > > task number when the slot number is the same. > > > > > > Giving an example to explain what's the difference between them: > > > > > > - A session cluster has 2 flink jobs, they are jobA and jobB > > > - Each TM has 4 slots. > > > - The task number of one slot of jobA is 3 > > > - The task number of one slot of jobB is 1 > > > - We have 2 TaskManagers: > > > - tm1 runs 3 slots of jobB, so tm1 runs 3 tasks > > > - tm2 runs 1 slot of jobA, and 1 slot of jobB, so tm2 runs 4 tasks. > > > > > > Now, we need to run a new slot, which tm should offer it? > > > - Option1: If we just check the task number, the tm1 is better. > > > - Option2: If we check the slot number first, and then check task, the > tm2 > > > is better > > > > > > The original FLIP selected option2, that's why we didn't add the > > > third option. The option2 didn't break the semantics when > > > `cluster.evenly-spread-out-slots` is true, and it just improve the > > > behavior without the semantics is changed. > > > > > > In the other hands, if we choose option2, when user set > > > `taskmanager.load-balance.mode` is Tasks. It also can achieve > > > the goal when it's Slots. > > > > > > So I think the `Slots` enum isn't needed if we choose option2. > > > Of course, If we choose the option1, the enum is needed. > > > > > > Looking forward to your feedback, thanks~ > > > > > > Best, > > > Rui > > > > > > On Wed, Sep 27, 2023 at 9:11 PM Zhu Zhu <reed...@gmail.com> wrote: > > > > > > > Thanks Yuepeng and Rui for creating this FLIP. > > > > > > > > +1 in general > > > > The idea is straight forward: best-effort gather all the slot > requests > > > > and offered slots to form an overview before assigning slots, trying > to > > > > balance the loads of task managers when assigning slots. > > > > > > > > I have one comment regarding the configuration for ease of use: > > > > > > > > IIUC, this FLIP uses an existing config > 'cluster.evenly-spread-out-slots' > > > > as the main switch of the new feature. That is, from user > perspective, > > > > with this improvement, the 'cluster.evenly-spread-out-slots' feature > not > > > > only balances the number of slots on task managers, but also > balances the > > > > number of tasks. This is a behavior change anyway. Besides that, it > also > > > > requires users to set 'slot.sharing-strategy' to > > > 'TASK_BALANCED_PREFERRED' > > > > to balance the tasks in each slot. > > > > > > > > I think we can introduce a new config option > > > > `taskmanager.load-balance.mode`, > > > > which accepts "None"/"Slots"/"Tasks". > `cluster.evenly-spread-out-slots` > > > > can be superseded by the "Slots" mode and get deprecated. In the > future > > > > it can support more mode, e.g. "CpuCores", to work better for jobs > with > > > > fine-grained resources. The proposed config option > > > > `slot.request.max-interval` > > > > then can be renamed to > > > > `taskmanager.load-balance.request-stablizing-timeout` > > > > to show its relation with the feature. The proposed > > > `slot.sharing-strategy` > > > > is not needed, because the configured "Tasks" mode will do the work. > > > > > > > > WDYT? > > > > > > > > Thanks, > > > > Zhu Zhu > > > > > > > > Yuepeng Pan <panyuep...@apache.org> 于2023年9月25日周一 16:26写道： > > > > > > > >> Hi all, > > > >> > > > >> > > > >> I and Fan Rui(CC’ed) created the FLIP-370[1] to support balanced > tasks > > > >> scheduling. > > > >> > > > >> > > > >> The current strategy of Flink to deploy tasks sometimes leads some > > > >> TMs(TaskManagers) to have more tasks while others have fewer tasks, > > > >> resulting in excessive resource utilization at some TMs that contain > > > more > > > >> tasks and becoming a bottleneck for the entire job processing. > > > Developing > > > >> strategies to achieve task load balancing for TMs and reducing job > > > >> bottlenecks becomes very meaningful. > > > >> > > > >> > > > >> The raw design and discussions could be found in the Flink JIRA[2] > and > > > >> Google doc[3]. We really appreciate Zhu Zhu(CC’ed) for providing > some > > > >> valuable help and suggestions in advance. > > > >> > > > >> > > > >> Please refer to the FLIP[1] document for more details about the > proposed > > > >> design and implementation. We welcome any feedback and opinions on > this > > > >> proposal. > > > >> > > > >> > > > >> [1] > > > >> > > > > https://cwiki.apache.org/confluence/display/FLINK/FLIP-370%3A+Support+Balanced+Tasks+Scheduling > > > >> > > > >> [2] https://issues.apache.org/jira/browse/FLINK-31757 > > > >> > > > >> [3] > > > >> > > > > https://docs.google.com/document/d/14WhrSNGBdcsRl3IK7CZO-RaZ5KXU2X1dWqxPEFr3iS8 > > > >> > > > >> > > > >> Best, > > > >> > > > >> Yuepeng Pan > > > >> > > > > > > > > > >
