Regulated enterprises often seek to extend their workloads into the cloud, but are impeded by integration concerns relating to security, governance and compliance. Further, enterprises running mission-critical applications, face throughput and latency challenges due to cloud integration overheads. We present Hybrid Cloud Connector to accelerate on-prem to cloud integration by handling non-functional aspects in lieu of the application, reducing complexity, and centralizing administration via a policy-driven control point.

The Sky vision[3] aims to open a new era in cloud computing. Sky abstracts clouds and dynamically use multiple clouds to optimize workload execution. This enable users to focus on their business logic, rather than interact with multiple clouds, and manually optimize performance and costs.

This work tackles an inherent conflict between two important trends. The first is the integration of data compression capabilities into many storage systems supporting random I/O on the compressed data. The second is encrypting data at the host, before data is written to the storage, in order to address regulatory and enterprise requirements. This provides end-to-end protection for the data, but since the data arrives encrypted, it prevents the storage from compressing the data. Can compression savings be achieved together with host side encryption without changing the storage protocols or storage backend? In this paper we show that they can.

The adoption of deduplication in storage systems has introduced significant new challenges for storage management. Specifically, the physical capacities associated with volumes are no longer readily available. In this work, we introduce a new approach to analyzing capacities in deduplicated storage environments. We provide sketch-based estimations of fundamental capacity measures required for managing a storage system: How much physical space would be reclaimed if a volume or group of volumes were to be removed from a system (the reclaimable capacity) and how much of the physical space should be attributed to each of the volumes in the system (the attributed capacity). Our methods also support capacity queries for volume groups across multiple storage systems, e.g., how much capacity would a volume group consume after being migrated to another storage system? We provide analytical accuracy guarantees for our estimations as well as empirical evaluations. Our technology is integrated into a prominent all-flash storage array and exhibits high performance even for very large systems. We also demonstrate how this method opens the door for performing placement decisions at the data-center level and obtaining insights on deduplication in the field.

In recent years, machine learning and deep learning techniques such as deep neural networks and recurrent neural networks have found uses in diverse fields including computer vision, speech recognition, natural language processing, social network analysis, bioinformatics and medicine, where they have produced results comparable to and in some cases surpassing human experts. Machine learning requires large amount of data for training its models with much of this data residing in object storage, an inexpensive and scalable data store. Also, deep learning make use of state of the art processing capabilities from high-end GPUs and accelerators, such as Google Tensor Processing Units (TPUs), which enable parallel and efficient execution. The throughput that such GPUs can support is very high.

Storage class memory (SCM) is an emerging class of memory devices that are both byte addressable, and persistent. There are many different technologies that can be considered SCM, at different stages of maturity. Examples of such technologies include NVDIMM-N, PCM, SttRAM, Racetrack, FeRAM, and others.

Cloud computing and virtualized infrastructures are currently the baseline environments for the provision of services in different application domains. While the number of service consumers increasingly grows, service providers aim at exploiting infrastructures that enable non-disruptive service provisioning, thus minimizing or even eliminating downtime. Nonetheless, to achieve the latter current approaches are either application-specific or cost inefficient, requiring the use of dedicated hardware. In this paper we present the reference architecture of a fault-tolerance scheme, which not only enhances cloud environments with the aforementioned capabilities but also achieves high-performance as required by mission critical every day applications. To realize the proposed approach, a new paradigm for memory and I/O externalization and consolidation is introduced, while current implementation references are also provided.

Reduction of disk drive power consumption is a challenging task, particularly since the most prevalent way of achieving it, powering down idle disks, has many undesirable side-effects. Some hard disk drives support acoustic modes, meaning they can be configured to reduce the acceleration and velocity of the disk head. This reduces instantaneous power consumption but sacrifices performance. As a result, input/output (I/O) operations run longer at reduced power. This is useful for power capping since it causes significant reduction in peak power consumption of the disks. We conducted experiments on several disk drives that support acoustic management. Most of these disk drives support only two modes - quiet and normal. We ran different I/O workloads, including SPC-1 to simulate a real-world online transaction processing workload. We found that the reduction in peak power can reach up to 23% when using quiet mode. We show that for some workloads this translates into a reduction of 12.5% in overall energy consumption. In other workloads we encountered the opposite phenomenon-an increase of more than 6% in the overall energy consumption.

Power consumption is a major issue in today's datacenters. Storage typically comprises a significant percentage of datacenter power. Thus, understanding, managing, and reducing storage power consumption is an essential aspect of any efforts that address the total power consumption of datacenters. We developed a scalable power modeling method that estimates the power consumption of storage workloads. The modeling concept is based on identifying the major workload contributors to the power consumed by the disk arrays.

Peer-to-peer systems are prone to faults, thus it is vitally important to design peer-to-peer systems to automatically regain consistency, namely to be self-stabilizing. Toward this goal, we present a deterministic structure that defines for every n the entire (IP) pointers structure among the n machines. Namely, the next hop for the insert, delete and search procedures of the peer-to-peer system. Thus, the consistency of the system is easily defined, monitored, verified and repaired. We present the HyperTree (distributed) structure which support the peer-to-peer procedures while ensuring that the out-degree and in-degree (the number of outgoing/incoming pointers) are b log/sub b/ N where N in the maximal number of machines and b is an integer parameter greater than 1. In addition the HyperTree ensures that the maximal number of hops involved in each procedure is bounded by log/sub b/ N. A self-stabilizing peer-to-peer system based on the HyperTree is presented.

A self-stabilizing distributed file system is presented. The system constructs and maintains a spanning tree for each file volume. The spanning tree consists of the servers that have volume replicas and caches for the specific file volume. The spanning trees are constructed and maintained by self-stabilizing distributed algorithms. File system updates use the tree to implement file read and write operations.