End users and businesses are increasingly relying on large-scale cloud computing systems to support Internet and telecommunication services such as e-Commerce, VoIP, business analytics etc., demanding quality of service (QoS) as well as quality of protection (QoP) guarantees from the system. Cloud providers typically use monitoring and management middleware to ensure that they can provide these guarantees to their end users.

The monitoring middleware needs to be able to detect failures when they occur in the system. Ideally the monitoring mechanisms should be capable of automatically inferring the root cause of any failures that occur in the system. In addition, system administrators often prefer better understanding of what might be going on in the system to make their own decisions besides relying on any automatic diagnosis provided by the monitoring middleware. In such cases the middleware must be able to generate effective visualizations that allow system administrators to quickly verify the results of the automatic diagnosis or to quickly infer an alternative root cause of the problem. In addition to these requirements, the monitoring middleware itself must be trustworthy and be capable of withstanding accidental as well as malicious faults. Building such a middleware is challenging due to the large amount of monitoring data that needs to be collected and analyzed, the necessity to recover and adapt in the presence of overloads or attacks, and ensuring the consistency of distributed and replicated monitoring activities.

Furthermore, the networking support needed for resilient communications that provide the above mentioned dependability and security guarantees for the monitoring and management middleware raises security issues that have not been addressed by the conventional networking community.

The purpose of this workshop is to bring together researchers and system administrators who share the common goals of managing a cloud computing system to discuss new ideas and techniques for building resilient and trustworthy monitoring/management middleware for such systems. Participants will discuss novel ways of architecting such a middleware, verifying the resilience and trustworthiness of the middleware, their experience with using such a middleware in cloud computing systems and other hot topics in the area of secure and trustworthy monitoring and management. The workshop will be a forum where networking researchers will meet the cloud monitoring/management community who share the common objectives of trustworthy network and service operations.

Topics of Interest

• Architectures for dependable, secure and trustworthy monitoring
• Techniques for analysis and visualization of system data
• Determination of malfunction or misuse based on system data
• Novel instrumentation sources and techniques for monitoring
• Intrusion tolerance in data centers and cloud infrastructures
• Dependability and security in virtualization technology
• Network architectures for the cloud monitoring
• Network resilience in cloud monitoring
• Software Defined Networks for trustworthy cloud monitoring
• Service tailored transport for resilience in the cloud
• Traffic and load balance monitoring in the cloud