In recent years, service mesh frameworks have gained significant popularity in building microservice-based applications. A key component of these frameworks is a proxy in each K8s pod, named sidecar, which handles inter-pod traffic. Our empirical measurement reveals that such per-pod sidecars cause numerous problems, including intrusion into the user pod, excessive resource occupation, significant overhead in managing many sidecars, and performance degradation caused by passing traffic through the sidecar.

Nowdays, it is prevalent to train deep learning models in cloud-native platforms that actively leverage containerization and orchestration technologies for high elasticity, low and flexible operation cost, and many other benefits. However, it also faces new challenges and our work is focusing on those related to I/O throughput for training, including complex data access, lack of matching dynamic I/O requirement, and inefficient I/O resource scheduling across different jobs. We proposeFluid, a cloud-native platform that provides DL training jobs with high-level data abstraction calledFluid Datasetto access training data from heterogeneous sources with elastic data acceleration. In addition, it comes with an on-the-fly cache system autoscaler that can match the online training speed and increase the number of cache replicas adaptively to alleviate I/O bottlenecks. To improve the overall performance of multiple DL jobs, Fluid co-orchestrate the data cache and DL jobs by arranging job scheduling in an appropriate order and can also schedule data cache and DL jobs on the same node to realize cache affinity. Experimental results show significant performance improvement of each individual DL job which uses dynamic computing resources with Fluid. For scheduling multiple DL jobs with same datasets, Fluid achieves around 2x performance speedup when integrated with existing widely-used and cutting-edge scheduling solutions through the appropriate job scheduling order. Besides, the cache affinity scheduling policy also improves job execution performance significantly. Fluid is now an open source project hosted by Cloud Native Computing Foundation (CNCF) with many production adopters.

In edge computing, the trend is moving towards leveraging cloud native technologies and platforms such as con-tainerization and Kubernetes to manage edge applications to improve operation efficiency. Unfortunately, the supported number of nodes per cluster is only several thousand in Kubernetes which is much less than what typical far edge use cases require. In this paper, we propose KOLE, a framework that extends the upstream Kubernetes for supporting a large number of far edge nodes. It replaces the existing apiserver-to-node communication mechanism in Kubernetes with a MQTT messaging system. The MQTT broker completely offloads the overhead of keeping numerous HTTP connections for nodes in the apiserver. In KOLE, we avoid creating numerous individual objects in the apiserver by maintaining them in a cloud state cache. The cache is snapshotted periodically for disaster recovery. Overall, KOLE achieves outstanding scalability by sacrificing the manageability of having individual objects, which we believe is a reasonable trade-off for far edge use cases. The experiment results show that KOLE is scalable and can support up to one million nodes.

Nowdays, it is prevalent to train deep learning (DL) models in cloud-native platforms that actively leverage containerization and orchestration technologies for high elasticity, low and flexible operation cost, and many other benefits. However, it also faces new challenges and our work is focusing on those related to I/O throughput for training, including complex data access with complicated performance tuning, lack of cache capacity with specialized hardware to match its high and dynamic I/O requirement, and inefficient I/O resource sharing across different training jobs. We propose Fluid, a cloud-native platform that provides DL training jobs with a data abstraction called Fluid Dataset to access training data from heterogeneous sources in a unified manner with transparent and elastic data acceleration powered by auto-tuned cache runtimes. In addition, it comes with an on-the-fly cache system autoscaler that can intelligently scale up and down the cache capacity to match the online training speed of each individual DL job. To improve the overall performance of multiple DL jobs, Fluid can co-orchestrate the data cache and DL jobs by arranging job scheduling in an appropriate order. Our experimental results show significant performance improvement of each individual DL job which uses dynamic computing resources with Fluid. In addition, for scheduling multiple DL jobs with same datasets, Fluid gives around 2x performance speedup when integrated with existing widely-used and cutting-edge scheduling solutions. Fluid is now an open source project hosted by Cloud Native Computing Foundation (CNCF) with adopters in production including Alibaba Cloud, Tencent Cloud, Weibo.com, China Telecom, etc.