SBIR Phase II: Providing Automatic System Anomaly Management Software as a Service for Dynamic Complex Computing Infrastructures

该小型企业创新研究 (SBIR) 第二阶段项目的更广泛影响/商业潜力将大大提高包括公共和私有计算云在内的许多计算基础设施的稳健性和可诊断性。所提出的技术将显着减少云计算基础设施中性能下降和服务停机的发生，从而可以吸引更多用户采用云计算技术，从而造福整个日益依赖云技术的社会。该项目还将通过将研究成果应用于现实世界来推进云系统可靠性研究的最先进水平。这个小型企业创新研究 (SBIR) 第二阶段项目将改变动态复杂计算基础设施的系统异常管理。该公司解决方案的新颖之处在于三个独特的功能：1）预测性：该解决方案可以在严重的服务中断发生之前发出预先警报； 2）自学习：解决方案自动推断警报条件并使用机器学习算法进行自动根本原因分析； 3）适应性：技术适应动态系统。拟议的研究将产生新颖且实用的异常预测和诊断解决方案，这些解决方案将在现实世界的计算基础设施中得到验证。具体来说，该项目包括三个主旨：1）动态环境中的自适应学习； 2）对系统指标和日志数据进行实时特征提取和模式识别； 3) 全栈根本原因分析。在项目期间，该公司将实施其软件产品，并与潜在客户就现实世界的计算基础设施进行案例研究。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project will be to greatly improve the robustness and diagnosability of many computing infrastructures including both public and private computing clouds. The proposed technology will significantly reduce the occurrence of performance degradation and service downtime in cloud computing infrastructures, which can attract more users to adopt cloud computing technology and thus benefit society as a whole, which depends increasingly on cloud technology. The project will also advance the state of the art in cloud system reliability research by putting research results into real world use. This Small Business Innovation Research (SBIR) Phase II project will transform system anomaly management for dynamic complex computing infrastructures. The novelty of the company's solution lies in three unique features: 1) predictive: the solution can raise advance alerts before a serious service outage occurs; 2) self-learning: the solution automatically infers alert conditions and performs automatic root cause analysis using machine learning algorithms; 3) adaptive: the technology adapts to dynamic systems. The proposed research will produce novel and practical anomaly prediction and diagnosis solutions that will be validated in real world computing infrastructures. Specifically, the project consists of three thrusts: 1) adaptive learning in dynamic environments; 2) real-time feature extraction and pattern recognition over system metric and log data; and 3) full stack root cause analysis. During the project the company will implement its software products and carry out case studies with prospective customers on real world computing infrastructures.