Cloud computing’s initial success and impact in the IT sector is undeniable, but as its service offerings continue to evolve, both cloud infrastructure providers and application developers alike look towards a significant improvement in terms of performance, reliability and agility.
Cloud providers (IaaS) look for ways to better optimize their infrastructure and drive down OpEx. Meanwhile, their customers, cloud application developers (SaaS), seek quicker time-to-market, shortened development/maintenance cycles and increased QoS/QoE for the end-user.
Holistic and cloud-specific engineering methods are needed to address the pain points of both stakeholders. Rising trends such as cloud-based monitoring and DevOps seek to better facilitate interactions between operations with development, but current solutions fall short in integrating such services between the infrastructure and application layers of the cloud.
CloudWave (www.cloudwave-fp7.eu) enables a next generation of cloud infrastructure operations and agile development by dynamically adapting cloud services to their environment, improving service quality and optimizing resource utilization. Inspired by the emerging concept of DevOps for facilitating application development, CloudWave empowers cloud infrastructure providers (IaaS) and their hosted applications developers (SaaS) to transparently collaborate to obtain high levels of service at lower costs. Execution Analytics: CloudWave improves existing cloud monitoring solutions with a more holistic and efficient approach towards IaaS and SaaS services. Unified monitoring consolidates infrastructure vs. application data, as well as virtual infrastructure vs. physical hardware. Programmable monitoring enhances filtering and delivery of data for analysis, allowing for better management of cloud-based resources. Coordinated Adaption: CloudWave enables reconfiguration of the infrastructure and application in real-time to compensate for a variety of performance factors, resulting in an increasingly resilient, automated and optimized cloud deployment. Feedback-driven Development: CloudWave advances current DevOps solutions with developer-oriented data based on its innovative monitoring. It mixes automation (coordinated adaptation) with customizable feedback for improved agile development, resulting in quicker time-to-market, shortened maintenance cycles and more reliable cloud applications for their end-user customers.
CloudWave's exploitation plan is two-pronged - there will be an internal exploitation of project results by the partner organizations themselves as well as external exploitaton of portions of the project, which include both as commercial offerings and as supported OpenSource. As an example of supported OpenSource, CloudWave is involved in the OpenStack Watcher initiative which is an incubator project for bringing new features and functionalities to the OpenStack core services. In that project, an architecture and optimization mechanism are being developed and CloudWave (through Intel’s participation) is transferring projects’ assets towards influencing the architecture and operation of Watcher. Specifically, two blueprints have been submitted and approved by the Watcher committee: ‘Mechanism for the consolidation of multiple actions plans to a single one’ (https://blueprints.launchpad.net/watcher/+spec/action-plans-consolidation) and ‘A graph model describing virtual and physical elements in a data center’ (https://blueprints.launchpad.net/watcher/+spec/graph-based-cluster-model). Both of them have been prioritized as inputs targeted for the next “Newton” Watcher release, planned for November 2016. The specifications of the two blueprints have been submitted to the Gerit Openstack review process (https://review.openstack.org/#/q/topic:bp/action-plans-consolidation,n,z, https://review.openstack.org/#/q/topic:bp/graph-based-cluster-model,n,z).
Open Source software available on