Eno Thereska

Migrating server storage to SSDs: analysis of tradeoffs

Recently, flash-based solid-state drives (SSDs) have become standard options for laptop and desktop storage, but their impact on enterprise server storage has not been studied. Provisioning server storage is challenging. It requires optimizing for the performance, capacity, power and reliability needs of the expected workload, all while minimizing financial costs. In this paper we analyze a number of workload traces from servers in both large and small data centers, to decide whether and how SSDs should be used to support each. We analyze both complete replacement of disks by SSDs, as well as use of SSDs as an intermediate tier between disks and DRAM. We describe an automated tool that, given device models and a block-level trace of a workload, determines the least-cost storage configuration that will support the workloads performance, capacity, and fault-tolerance requirements. We found that replacing disks by SSDs is not a costeffective option for any of our workloads, due to the low capacity per dollar of SSDs. Depending on the workload, the capacity per dollar of SSDs needs to increase by a factor of 3-3000 for an SSD-based solution to break even with a diskbased solution. Thus, without a large increase in SSD capacity per dollar, only the smallest volumes, such as system boot volumes, can be cost-effectively migrated to SSDs. The benefit of using SSDs as an intermediate caching tier is also limited: fewer than 10% of our workloads can reduce provisioning costs by using an SSD tier at todays capacity per dollar, and fewer than 20% can do so at any SSD capacity per dollar. Although SSDs are much more energy-efficient than enterprise disks, the energy savings are outweighed by the hardware costs, and comparable energy savings are achievable with low-power SATA disks.

Dynamic right-sizing for power-proportional data centers

Power consumption imposes a significant cost for data centers implementing cloud services, yet much of that power is used to maintain excess service capacity during periods of predictably low load. This paper investigates how much can be saved by dynamically ‘right-sizing’ the data center by turning off servers during such periods, and how to achieve that saving via an online algorithm. We prove that the optimal offline algorithm for dynamic right-sizing has a simple structure when viewed in reverse time, and this structure is exploited to develop a new ‘lazy’ online algorithm, which is proven to be 3-competitive. We validate the algorithm using traces from two real data center workloads and show that significant cost-savings are possible.

Sierra: practical power-proportionality for data center storage

Online services hosted in data centers show significant diurnal variation in load levels. Thus, there is significant potential for saving power by powering down excess servers during the troughs. However, while techniques like VM migration can consolidate computational load, storage state has always been the elephant in the room preventing this powering down. Migrating storage is not a practical way to consolidate I/O load. This paper presents Sierra, a power-proportional distributed storage subsystem for data centers. Sierra allows powering down of a large fraction of servers during troughs without migrating data and without imposing extra capacity requirements. It addresses the challenges of maintaining read and write availability, no performance degradation, consistency, and fault tolerance for general I/O workloads through a set of techniques including power-aware layout, a distributed virtual log, recovery and migration techniques, and predictive gear scheduling. Replaying live traces from a large, real service (Hotmail) on a cluster shows power savings of 23%. Savings of 40--50% are possible with more complex optimizations.

Stardust: tracking activity in a distributed storage system

Performance monitoring in most distributed systems provides minimal guidance for tuning, problem diagnosis, and decision making. Stardust is a monitoring infrastructure that replaces traditional performance counters with end-to-end traces of requests and allows for efficient querying of performance metrics. Such traces better inform key administrative performance challenges by enabling, for example, extraction of per-workload, per-resource demand information and per-workload latency graphs. This paper reports on our experience building and using end-to-end tracing as an on-line monitoring tool in a distributed storage system. Using diverse system workloads and scenarios, we show that such fine-grained tracing can be made efficient (less than 6% overhead) and is useful for on- and off-line analysis of system behavior. These experiences make a case for having other systems incorporate such an instrumentation framework.

Lost in translation: understanding the possession of digital things in the cloud

People are amassing larger and more diverse collections of digital things. The emergence of Cloud computing has enabled people to move their personal files to online places, and create new digital things through online services. However, little is known about how this shift might shape peoples orientations toward their digital things. To investigate, we conducted in depth interviews with 13 people comparing and contrasting how they think about their possessions, moving from physical ones, to locally kept digital materials, to the online world. Findings are interpreted to detail design and research opportunities in this emerging space.

File classification in self-* storage systems

To tune and manage themselves, file and storage systems must understand key properties (e.g., access pattern, lifetime, size) of their various files. This paper describes how systems can automatically learn to classify the properties of files (e.g., read-only access pattern, short-lived, small in size) and predict the properties of new files, as they are created, by exploiting the strong associations between a files properties and the names and attributes assigned to it. These associations exist, strongly but differently, in each of four real NFS environments studied. Decision tree classifiers can automatically identify and model such associations, providing prediction accuracies that often exceed 90%. Such predictions can be used to select storage policies (e.g., disk allocation schemes and replication factors) for individual files. Further, changes in associations can expose information about applications, helping autonomic system components distinguish growth from fundamental change.

Towards self-predicting systems: What if you could ask ‘what-if’?

Today, management and tuning questions are approached using if...then... rules of thumb. This reactive approach requires expertise regarding of system behavior, making it difficult to deal with unforeseen uses of a system’s resources and leading to system unpredictability and large system management overheads. We propose a What...if... approach that allows interactive exploration of the effects of system changes, thus converting complex tuning problem into simpler search problems. Through two concrete management problems, automating system upgrades and deciding on service migrations, we identify system design changes that enable a system to answer What...if... ques-

Software-defined caching: managing caches in multi-tenant data centers

In data centers, caches work both to provide low IO latencies and to reduce the load on the back-end network and storage. But they are not designed for multi-tenancy; system-level caches today cannot be configured to match tenant or provider objectives. Exacerbating the problem is the increasing number of un-coordinated caches on the IO data plane. The lack of global visibility on the control plane to coordinate this distributed set of caches leads to inefficiencies, increasing cloud provider cost. We present Moirai, a tenant- and workload-aware system that allows data center providers to control their distributed caching infrastructure. Moirai can help ease the management of the cache infrastructure and achieve various objectives, such as improving overall resource utilization or providing tenant isolation and QoS guarantees, as we show through several use cases. A key benefit of Moirai is that it is transparent to applications or VMs deployed in data centers. Our prototype runs unmodified OSes and databases, providing immediate benefit to existing applications.

Non-static nature of patient consent: shifting privacy perspectives in health information sharing

The purpose of the study is to explore how chronically ill patients and their specialized care network have viewed their personal medical information privacy and how it has impacted their perspectives of sharing their records with their network of healthcare providers and secondary use organizations. Diabetes patients and specialized diabetes medical care providers in Eastern England were interviewed about their sharing of medical information and their privacy concerns to inform a descriptive qualitative and exploratory thematic analysis. From the interview data, we see that diabetes patients shift their perceived privacy concerns and needs throughout their lifetime due to persistence of health data, changes in health, technology advances, and experience with technology that affect ones consent decisions. From these findings, we begin to take a translational research approach in critically examining current privacy enhancing technologies for secondary use consent management and motivate the further exploration of both temporally-sensitive privacy perspectives and new options in consent management that support shifting privacy concerns over ones lifetime.

Towards Self-Predicting Systems: What If You Could Ask "What-If"?

Today, management and tuning questions are approached using if...then... rules of thumb. This reactive approach requires expertise regarding of system behavior, making it difficult to deal with unforeseen uses of a systems resources and leading to system unpredictability and large system management overheads. We propose a what...if... approach that allows interactive exploration of the effects of system changes, thus converting complex tuning problem into simpler search problems. Through two concrete management problems, automating system upgrades and deciding on service migrations, we identify system design changes that enable a system to answer what...if... questions about itself