Root cause analysis (RCA) is critical for maintaining the availability and efficiency of cloud computing systems. However, identifying root causes from the large-scale, high-dimensional monitoring data generated by these complex environments is a significant challenge. Current approaches often rely on time-consuming manual analysis to ensure flexibility and reliability, while recent automated methods lack the crucial insights provided by domain experts. To bridge this gap, we propose RCInvestigator, a visual analytics system that facilitates interactive root cause investigation by establishing a tight collaboration between human experts and machine analysis. Our approach addresses three key challenges: a) modeling databases for the root cause investigation, b) inferring root causes from large-scale time series, and c) building comprehensible investigation results. We demonstrate the effectiveness and utility of RCInvestigator through two real-world case studies, which received positive feedback from domain experts.

In virtual machine (VM) allocation systems, caching repetitive and similar VM allocation requests and associated resolution rules is crucial for reducing computational costs and meeting strict latency requirements. While modern allocation systems distribute requests among multiple allocator agents and use caching to improve performance, current schedulers often neglect the cache state and latency considerations when assigning each new request to an agent. Due to the high variance in costs of cache hits and misses and the associated processing overheads of updating the caches, simple load-balancing and cache-aware mechanisms result in high latencies. We introduce Kamino, a high-performance, latency-driven and cache-aware request scheduling system aimed at minimizing end-to-end latencies. Kamino employs a novel scheduling algorithm grounded in theory which uses partial indicators from the cache state to assign each new request to the agent with the lowest estimated latency. Evaluation of Kamino using a high-fidelity simulator on large-scale production workloads shows a 42% reduction in average request latencies. Our deployment of Kamino in the control plane of a large public cloud confirms these improvements, with a 33% decrease in cache miss rates and 17% reduction in memory usage.

The advent of abundant image data has catalyzed the advancement of visual control in reinforcement learning (RL) systems, leveraging multiple view- points to capture the same physical states, which could enhance control performance theoretically. However, integrating multi-view data into representation learning remains challenging. In this paper, we introduce SMuCo, an innovative multi-view reinforcement learning algorithm that constructs robust latent representations by optimizing multi- view sequential total correlation. This technique effectively captures task-relevant information and temporal dynamics while filtering out irrelevant data. Our method supports an unlimited number of views and demonstrates superior performance over leading model-free and model-based RL algorithms. Empirical results from the DeepMind Control Suite and the Sapien Basic Manipulation Task confirm SMuCo’s enhanced efficacy, significantly improving task performance across diverse scenarios and views.

Efficient resource utilization and perfect user experience usually conflict with each other in cloud computing platforms. Great efforts have been invested in increasing resource utilization but trying not to affect users' experience for cloud computing platforms. In order to better utilize the remaining pieces of computing resources spread over the whole platform, deferrable jobs are provided with a discounted price to users. For this type of deferrable jobs, users are allowed to submit jobs that will run for a specific uninterrupted duration in a flexible range of time in the future with a great discount. With these deferrable jobs to be scheduled under the remaining capacity after deploying those on-demand jobs, it remains a challenge to achieve high resource utilization and meanwhile shorten the waiting time for users as much as possible in an online manner. In this paper, we propose an online deferrable job scheduling method called Online Scheduling for DEferrable jobs in Cloud (), where a deep reinforcement learning model is adopted to learn the scheduling policy, and several auxiliary tasks are utilized to provide better state representations and improve the performance of the model. With the integrated reinforcement learning framework, the proposed method can well plan the deployment schedule and achieve a short waiting time for users while maintaining a high resource utilization for the platform. The proposed method is validated on a public dataset and shows superior performance.

Quota is often used in resource allocation and management scenarios to prevent abuse of resource and increase the efficiency of resource utilization. Quota management is usually fulfilled with a set of rules maintained by the system administrator. However, maintaining these rules usually needs deep domain knowledge. Moreover, arbitrary rules usually cannot guarantee both high resource utilization and fairness at the same time. In this paper, we propose a reinforcement learning framework to automatically respond to quota requests in cloud computing platforms with distinctive usage characteristics for users. Extensive experimental results have demonstrated the superior performance of our framework on achieving a great trade-off between efficiency and fairness.

Offline reinforcement learning faces a significant challenge of value over-estimation due to the distributional drift between the dataset and the current learned policy, leading to learning failure in practice. The common approach is to incorporate a penalty term to reward or value estimation in the Bellman iterations. Meanwhile, to avoid extrapolation on out-of-distribution (OOD) states and actions, existing methods focus on conservative Q-function estimation. In this paper, we propose Conservative State Value Estimation (CSVE), a new approach that learns conservative V-function via directly imposing penalty on OOD states. Compared to prior work, CSVE allows more effective state value estimation with conservative guarantees and further better policy optimization. Further, we apply CSVE and develop a practical actor-critic algorithm in which the critic does the conservative value estimation by additionally sampling and penalizing the states around the dataset, and the actor applies advantage weighted updates extended with state exploration to improve the policy. We evaluate in classic continual control tasks of D4RL, showing that our method performs better than the conservative Q-function learning methods and is strongly competitive among recent SOTA methods.

Offline RL promises to learn effective policies from static experience datasets without further interaction, which expect to perform well in the online environment. However, it faces up to a major challenge of value over-estimation introduced by the distributional drift between the dataset and the current learned policy, which leads to learning failure in practice. The common approach is to add a penalty term to reward or value estimation in the Bellman iterations, which has given rise to a number of successful algorithms such as CQL. Meanwhile, to avoid extrapolation on unseen states, existing methods focus on conservative Q-function estimation. In this paper, we propose CSVE, a new approach that directly imposes penalty on out-of-distribution states. We prove that for the evaluated policy, our conservative state value estimation satisfies: (1) over the state distribution that samples penalizing states, it lower bounds the true values in expectation, and (2) over the marginal state distribution of data, it is no more than the true values in expectation plus a constant decided by sampling error. Further, we develop a practical actor-critic algorithm in which the critic does the conservative value estimation by additionally sampling and penalizing the states 'around' the dataset, while the actor applies advantage weighted updates to improve the policy. We evaluate in classic continual control tasks of D4RL, showing that our method performs better than the conservative Q-function learning methods (e.g., CQL) and is strongly competitive among recent SOTA methods.

The infinite capacity of cloud computing is an illusion: in reality, cloud providers cannot always have enough capacity of the right type, in the right place, at the right time to meet all demand. Consequently, cloud providers need to implement admission-control policies to ensure accepted capacity requests experience high availability. However, admission control in the public cloud is hard due to dynamic changes in both supply and demand: hardware might become unavailable, and actual VM consumption could vary for a variety of reasons including tenant scale-outs and fulfillment of VM reservations made by customers ahead of time. In this paper, we design and implement Kerveros, a flexible admission-control system that has three desired properties: i) high computational scalability to handle a large inventory, ii) accurate capacity provisioning for high VM availability, and iii) good packing efficiency to optimize resource usage. To achieve this, Kerveros uses novel bookkeeping techniques to quickly estimate the capacity available for incoming VM requests. Our system has been deployed in Microsoft Azure. Results from both simulations and production confirm that Kerveros achieves more than four nines of availability while sustaining request processing latencies of a few milliseconds.

This paper investigates a critical resource allocation problem in the first party cloud: scheduling containers to machines. There are tens of services, and each service runs a set of homogeneous containers with dynamic resource usage; containers of a service are scheduled daily in a batch fashion. This problem can be naturally formulated as Stochastic Bin Packing Problem (SBPP). However, traditional SBPP research often focuses on cases of empty machines, whose objective, i.e., to minimize the number of used machines, is not well-defined for the more common reality with nonempty machines. This paper aims to close this gap. First, we define a new objective metric, Used Capacity at Confidence (UCaC), which measures the maximum used resources at a probability and is proved to be consistent for both empty and nonempty machines and reformulate the SBPP under chance constraints. Second, by modeling the container resource usage distribution in a generative approach, we reveal that UCaC can be approximated with Gaussian, which is verified by trace data of real-world applications. Third, we propose an exact solver by solving the equivalent cutting stock variant as well as two heuristics-based solvers -- UCaC best fit, bi-level heuristics. We experimentally evaluate these solvers on both synthetic datasets and real application traces, demonstrating our methodology's advantage over traditional SBPP optimal solver minimizing the number of used machines, with a low rate of resource violations.

Capacity management has always been a great challenge for cloud platforms due to massive, heterogeneous on-demand instances running at different times. To better plan the capacity for the whole platform, a class of cloud computing instances have been released to collect computing demands beforehand. To use such instances, users are allowed to submit jobs to run for a pre-specified uninterrupted duration in a flexible range of time in the future with a discount compared to the normal on-demand instances. Proactively scheduling those pre-collected job requests considering the capacity status over the platform can greatly help balance the computing workloads along time. In this work, we formulate the scheduling problem for these pre-collected job requests under uncertain available capacity as a Prediction + Optimization problem with uncertainty in constraints, and propose an effective algorithm called Controlling under Uncertain Constraints (CUC), where the predicted capacity guides the optimization of job scheduling and job scheduling results are leveraged to improve the prediction of capacity through Bayesian optimization. The proposed formulation and solution are commonly applicable for proactively scheduling problems in cloud computing. Our extensive experiments on three public, industrial datasets shows that CUC has great potential for supporting high reliability in cloud platforms.

In cloud systems, an accurate capacity planning is very important for cloud provider to improve service availability. Traditional methods simply predicting "when the available resources is exhausted" are not effective due to customer demand fragmentation and platform allocation constraints. In this paper, we propose a novel prediction approach which proactively predicts the level of resource allocation failures from the perspective of low capacity status. By jointly considering the data from different sources in both time series form and static form, the proposed approach can make accurate LCS predictions in a complex and dynamic cloud environment, and thereby improve the service availability of cloud systems. The proposed approach is evaluated by real-world datasets collected from a large scale public cloud platform, and the results confirm its effectiveness.

For proof-of-work blockchains such as Ethereum, the mining power decentralization is an important discussion point in the community. Previous studies mostly focus on the aggregated power of the mining pools, neglecting the pool participants who are the source of the pools' power. In this paper, we present the first large-scale study of the pool participants in Ethereum's mining pools. Pool participants are not directly observable because they communicate with their pools via private channels. However, they leave "footprints" on chain as they use Ethereum accounts to anonymously receive rewards from mining pools. For this study, we combine several data sources to identify 62,358,646 pool reward transactions sent by 47 pools to their participants over Ethereum's entire near 5-year history. Our analyses about these transactions reveal interesting insights about three aspects of pool participants: the power decentralization at the participant level, their pool-switching behavior, and why they participate in pools. Our results provide a complementary and more balanced view about Ethereum's mining power decentralization at a deeper level.

The use of containers in microservices has gained popularity as it facilitates agile development, resource governance, and software maintenance. Container reallocation aims to achieve workload balance via reallocating containers over physical machines. It affects the overall performance of microservice-based systems. However, container scheduling and reallocation remain an open issue due to their complexity in real-world scenarios. In this paper, we propose a novel Multi-Phase Local Search (MPLS) algorithm to optimize container reallocation. The experimental results show that our optimization algorithm outperforms state-of-the-art methods. In practice, it has been successfully applied to Microsoft 365 system to mitigate hotspot machines and balance workloads across the entire system.

We describe the design and implementation of Protean - the Microsoft Azure service responsible for allocating Virtual Machines (VMs) to millions of servers around the globe. A single instance of Protean serves an entire availability zone (10-100k machines), facilitating seamless failover and scale-out to customers. The design has proven robust, enabling a substantial expansion of VM offerings and features with minimal changes to the core infrastructure. In particular, Protean preserves a clear separation between policy and mechanisms. From a policy perspective, a flexible rule-based Allocation Agent (AA) allows Protean to efficiently address multiple constraints and performance criteria, and adapt to different conditions. On the system side, a multi-layer caching mechanism expedites the allocation process, achieving turnaround times of few milliseconds. A slight compromise on allocation quality enables multiple AAs to run concurrently on the same inventory, resulting in increased throughput with negligible conflict rate. Our results from both simulations and production demonstrate that Protean achieves high throughput and utilization (85-90% on a key utilization metric), while satisfying user-specific requirements. We also demonstrate how Protean is adapted to handle capacity crunch conditions, by zooming in on spikes caused by COVID-19.

Exploring the opportunities to use ML, the possible designs, and our experience with Microsoft Azure.

Services that rely heavily on machine learning (ML), such as speech understanding and image recognition, have been receiving significant attention. ML can also be used to optimize the cloud platforms underlying these services. The challenge is defining exactly how we should integrate ML into these platforms. In this paper, we motivate this integration, discuss multiple dimensions associated with it and the architectural tradeoffs in each dimension, and describe our first approach to building ML-centric cloud platforms. Specifically, as one point in the multi-dimensional space, we overview the Resource Central ML and prediction-serving system, its integration into the Microsoft Azure Compute fabric, and the lessons we learned. We conclude with a discussion of future research avenues.

Cloud providers rent the resources they do not allocate as evictable virtual machines (VMs), like spot instances. In this paper, we first characterize the unallocated resources in Microsoft Azure, and show that they are plenty but may vary widely over time and across servers. Based on the characterization, we propose a new class of VM, called Harvest VM, to harvest and monetize the unallocated resources. A Harvest VM is more flexible and efficient than a spot instance, because it grows and shrinks according to the amount of unallocated resources at its underlying server; it is only evicted/killed when the provider needs its minimum set of resources. Next, we create models that predict the availability of the unallocated resources for Harvest VM deployments. Based on these predictions, we provide Service Level Objectives (SLOs) for the survival rate (e.g., 65% of the Harvest VMs will survive more than a week) and the average number of cores that can be harvested. Our short-term predictions have an average error under 2% and less than 6% for longer terms. We also extend a popular cluster scheduling framework to leverage the harvested resources. Using our SLOs and framework, we can offset the rare evictions with extra harvested cores and achieve the same computational power as regular-priority VMs, but at 91% lower cost. Finally, we outline lessons and results from running Harvest VMs and our framework in production.

The present disclosure relates to systems, methods, and computer readable media for predicting deployment growth on one or more node clusters and selectively permitting deployment requests on a per cluster basis. For example, systems disclosed herein may apply tenant growth prediction system trained to output a deployment growth classification indicative of a predicted growth of deployments on a node cluster. The system disclosed herein may further utilize the deployment growth classification to determine whether a deployment request may be permitted while maintaining a sufficiently sized capacity buffer to avoid deployment failures for existing deployments previously implemented on the node cluster. By selectively permitting or denying deployments based on a variety of factors, the systems described herein can more efficiently utilize cluster resources on a per-cluster basis without causing a significant increase in deployment failures for existing customers.

FPGAs have been deployed at massive scale in data centers. Using currently available communication architectures, however, it is difficult for FPGAs to access and utilize the various heterogenous resources available in data centers (DRAM, CPU, GPU,...). In this paper, we present Direct Universal Access (DUA), a communication architecture that provides uniform access for FPGA to these data center resources. Without being limited by machine boundaries, DUA provides global names and a common interface for communicating across various resources, the underlying network automatically routing traffic and managing resource multiplexing. Our benchmarks show that DUA provides simple and fair-share resource access with small logic area overhead (< 10%) and negligible latency (< 0.2 mu s). We also build two practical multi-FPGA applications-deep crossing and regular expression matching-on top of DUA to demonstrate its usability and efficiency.