mpm + control-theory   3

Performance-Feedback Autoscaling with Budget Constraints for Cloud-based Workloads of Workflows
The growing popularity of workflows in the cloud domain promoted the development of sophisticated autoscaling policies that allow automatic allocation and deallocation of resources. However, many state-of-the-art autoscaling policies for workflows are mostly plan-based or designed for batches (ensembles) of workflows. This reduces their flexibility when dealing with workloads of workflows, as the workloads are often subject to unpredictable resource demand fluctuations. Moreover, autoscaling in clouds almost always imposes budget constraints that should be satisfied. The budget-aware autoscalers for workflows usually require task runtime estimates to be provided beforehand, which is not always possible when dealing with workloads due to their dynamic nature. To address these issues, we propose a novel Performance-Feedback Autoscaler (PFA) that is budget-aware and does not require task runtime estimates for its operation. Instead, it uses the performance-feedback loop that monitors the average throughput on each resource type. We implement PFA in the popular Apache Airflow workflow management system, and compare the performance of our autoscaler with other two state-of-the-art autoscalers, and with the optimal solution obtained with the Mixed Integer Programming approach. Our results show that PFA outperforms other considered online autoscalers, as it effectively minimizes the average job slowdown by up to 47% while still satisfying the budget constraints. Moreover, PFA shows by up to 76% lower average runtime than the competitors.
control-theory  scalability 
9 weeks ago by mpm
Worry-Free Ingestion: Flow Control of Writes in Scylla
We would like the ingestion to proceed as quickly as possible but without overwhelming the servers. An over-eager client may send write requests faster than the cluster can complete earlier requests. If this is only a short burst of requests, Scylla can absorb the excess requests in a queue or numerous queues distributed throughout the cluster (we’ll look at the details of these queues below). But had we allowed the client to continue writing at this excessive rate, the backlog of uncompleted writes would continue to grow until the servers run out of memory and possibly crash. So as the backlog grows, we need to find a way for the server to tell the client to slow down its request rate. If we can’t slow down the client, we have to start failing new requests.
control-theory  load-balancing 
december 2018 by mpm
Taming the Beast: How Scylla Leverages Control Theory to Keep Compactions Under Control
We borrow from the mathematical framework of industrial controllers to make sure that compaction bandwidth is automatically set to a fair value while maintaining a predictable system response.
july 2018 by mpm

Copy this bookmark: