Benjamin Yolken

Using a market economy to provision compute resources across planet-wide clusters

We present a practical, market-based solution to the resource provisioning problem in a set of heterogeneous resource clusters. We focus on provisioning rather than immediate scheduling decisions to allow users to change long-term job specifications based on market feedback. Users enter bids to purchase quotas, or bundles of resources for long-term use. These requests are mapped into a simulated clock auction which determines uniform, fair resource prices that balance supply and demand. The reserve prices for resources sold by the operator in this auction are set based on current utilization, thus guiding the users as they set their bids towards under-utilized resources. By running these auctions at regular time intervals, prices fluctuate like those in a real-world economy and provide motivation for users to engineer systems that can best take advantage of available resources. These ideas were implemented in an experimental resource market at Google. Our preliminary results demonstrate an efficient transition of users from more congested resource pools to less congested resources. The disparate engineering costs for users to reconfigure their jobs to run on less expensive resource pools was evidenced by the large price premiums some users were willing to pay for more expensive resources. The final resource allocations illustrated how this framework can lead to significant, beneficial changes in user behavior, reducing the excessive shortages and surpluses of more traditional allocation methods.

Security Decision-Making among Interdependent Organizations

In various settings, such as when customers use the same passwords at several independent web sites, security decisions by one organization may have a significant impact on the security of another. We develop a model for security decision-making in such settings, using a variation of linear influence networks. The linear influence model uses a matrix to represent linear dependence between security investment at one organization and resulting security at another, and utility functions to measure the overall benefit to each organization. A simple matrix condition implies the existence and uniqueness of Nash equilibria, which can be reached by a natural iterative algorithm. A free-riding index, expressible using quantities computed in this model, measures the degree to which one organization can potentially reduce its security investment and benefit from investments of others. We apply this framework to investigate three examples: web site security with shared passwords, customer education against phishing and identity theft, and anti-spam email filters. While we do not have sufficient quantitative data to draw quantitative conclusions about any of these situations, the model provides qualitative information about each example.

Security investment games of interdependent organizations

In various computer security settings, such as when customers use the same passwords at several independent Web sites, security decisions made by one organization may have significant impact on the security of another. We develop a model for security decision-making in inter-dependent organizations described by a linear influence network. In this model, a matrix represents how one organizations investments are augmented by some linear function of its neighbors investments. Each element of the matrix, representing the strength of influence of one organization on another, can be positive or negative and need not be symmetric with respect to two organizations. A simple matrix condition implies the existence and uniqueness of Nash equilibria, which can be reached by a natural iterative algorithm. We demonstrate that there are ways of improving the matrix such that two organizations decrease their investments while all others maintain the same level of investment. We apply this framework to the setting of Web site security with shared passwords.

Game based capacity allocation for utility computing environments

Utility computing has the potential to greatly increase the efficiency of IT operations by sharing resources across multiple users. This sharing, however, introduces complex problems with regards to pricing and allocating these resources in a way that is fair, easy to implement, and economically efficient. In this paper, we study a queue-based model that attempts to address these issues. Each client/user has a continuous flow of jobs that need to be processed. The service rate each receives, however, is proportional to a bid it submits to the system operator. Assuming that user costs are some function of their average backlogs plus their bid amounts, we use this allocation mechanism to construct an economic game.Much previous research has shown that these types of allocation games have desirable properties if the cost functions are well-defined and convex over the space of possible outcomes. Because of its queueing interface, however, our model induces functions that do not satisfy the latter, commonly assumed properties. In spite of these complications, we show that the game still has a unique equilibrium and that the system will converge to this point if users iteratively make “best response” updates to their bids. Finally, we explore the “price of anarchy” in our model, proving a bound on efficiency losses as a function of several fundamental system parameters. Thus, our scheme results in equilibria with a number of highly desirably properties.

Power Aware Management of Packet Switches

In this paper, we extend the idea of dynamic power management (DPM) to network packet switches, devices whose increasing speeds and densities are leading to costly and physically cumbersome power dissipation problems. In particular, we take a system-level approach, examining how operations in an input-queued (IQ) switch can be scheduled to balance power consumption on the one hand with average delay on the other. To accomplish this, we formulate a quadratic cost model which accounts for the power / delay tradeoff and then pose the scheduling problem in terms of a dynamic program. Using techniques from the field of linear quadratic regulation (LQR), we then solve for the optimal control in a relaxed system model and use this to create a novel scheduling algorithm: power aware maximum weight matching (PAMWM). Initial simulation results suggest that, under low load conditions, our algorithm results in significant power savings compared to MWM, with little performance degradation.

Power Managed Packet Switching

High power dissipation in packet switches and routers is fast turning into a key problem, owing to increasing line speeds and decreasing chip sizes. To address this issue, we introduce and develop the notion of a power-managed input-queued (PMIQ) switch in this paper. A PMIQ switch is an input-queued switch with an additional hierarchy of control to regulate the power dissipated by the switch. We formulate the joint scheduling and power management problem for a PMIQ switch as a dynamic program (DP). Leveraging intuition gained from provable structural properties of the optimal solution to the DP, we propose the power-aware switch scheduling (PASS) switch management policy. PASS dynamically selects the rate/speed at which the switch operates, in conjunction with the switch configuration, as a function of the backlogs of the input buffers. PASS can easily be tuned to tradeoff high power consumption for larger queuing delays. Experimental results show that PASS yields an attractive power-delay tradeoff relative to the benchmark maximum weight matching (MWM) scheduler. The low computational complexity of PASS makes it amenable to implementation in large, high-speed switches.

Power and Delay Aware Management of Packet Switches

Due to increasing circuit densities and data throughput rates, power consumption has become a significant concern in the design and operation of high-performance packet switches. We extend the idea of Dynamic Power Management (DPM) to input queued switches, allowing operators to tradeoff power and delay in a useful way. We frame the problem as a dynamic program and solve a relaxation using techniques from Linear Quadratic Regulation (LQR). This optimal policy is combined with existing, nonpower-aware switch controls to generate two novel scheduling algorithms: 1) LQR Power Aware Maximum Weight Matching (LQR PA MWM) and 2) LQR Power Aware Projective Cone Scheduling (LQR PA PCS). Simulation results suggest that our algorithms result in significant power savings compared to MWM and previous power control schemes with little performance degradation.

Game-based admission control for wireless systems

Much previous work has examined the wireless power control problem using tools from game theory, an economic concept which describes the behavior of interdependent but non-cooperative users. In this paper, we expand these ideas to the antecedent process of deciding which users may participate in the network, i.e. the admission control problem. In particular, we propose three distinct pricing schemes for influencing users as they make their participation decisions. We fully characterize the equilibria induced by each and then test their performance in a simulated, wireless environment. Our preliminary results show that these schemes have the potential to produce high quality outcomes in an incentive-compatible way.

Backlog Aware Scheduling for Large Buffered Crossbar Switches

A novel architecture was proposed in [1] to address scalability issues in large, high speed packet switches. The architecture proposed in [1], namely OBIG (output buffers with input groups), distributes the switch fabric across multiple chips, which communicate via high speed interconnects enabled by proximity communication (PC), a recently developed circuit technology [2]. An OBIG switch aggregates multiple input flows inside the switch fabric, thereby significantly reducing the amount of memory required for internal buffers, vis-a-vis a conventional buffered crossbar, which has buffers at every crosspoint. Thus, the OBIG architecture is promising for realizing terabit switches with hundreds of ports. This paper studies packet scheduling algorithms which help realize the potential of OBIG-like switch architectures. The emphasis here is on designing backlog aware scheduling algorithms, while ensuring desirable traits such as low computational complexity and scalability. The efficacy of the proposed scheduling algorithms with respect to performance metrics such as average delay and fairness is demonstrated via simulations under a variety of scenarios.

Target-Based Power Control for Queueing Systems with Applications to Packet Switches

Many data center devices, for instance packet switches, can be modeled within the context of resource constrained queueing systems. In this paper, we define a novel algorithm class which simultaneously addresses three significant concerns in the operation of such systems: stability, differentiated QOS, and power control. This class, which we refer to as target/power projective cone scheduling (TP-PCS), encapsulates many previously studied algorithms as special cases. At the same time, however, it is broad enough to include a rich set of other, potentially superior control procedures. In the first part of our paper, we explain our model as well as some previously studied approaches to scheduling in these systems. We then define TP-PCS, show how it relates to the former algorithms, and discuss how members of this class can be tailored to control for both power and QOS. Finally, we test some instances of TP-PCS on a simulated, input-queued switch. These show that a wide variety of operating modes are possible by adjusting various scheduling parameters. Hence, TP-PCS opens up for exploration a large set of new controls for packet switches and other systems operating in the queueing space.