An official website of the United States government
Resource control in heterogeneous computers built with subsystems from different vendors is challenging. There is a tension between the need to quickly generate local decisions in each subsystem and the desire to coordinate the different subsystems for global optimization. In practice, global coordination among subsystems is considered hard, and current commercial systems use centralized controllers. The result is high response time and high design cost due to lack of modularity. To control emerging heterogeneous computers effectively, we propose a new control framework called Tangram that is fast, glob- ally coordinated, and modular. Tangram introduces a new formal controller that combines multiple engines for optimization and safety, and has a standard interface. Building the controller for a subsystem requires knowing only about that subsystem. As a het- erogeneous computer is assembled, the controllers in the different subsystems are connected hierarchically, exchanging standard co- ordination signals. To demonstrate Tangram, we prototype it in a heterogeneous server that we assemble using components from multiple vendors. Compared to state-of-the-art control, Tangram re- duces, on average, the execution time of heterogeneous applications by 31% and their energy-delay product by 39%.
more »
« less
Structured Singular Value Control for Modular Resource Management in Multilayer Computers
Computer systems are operating in environments where applications are rapidly diversifying while resources like energy and storage are becoming severely limited. These environments demand that computers dynamically manage their resources efciently to deliver the best performance and meet many goals. An important challenge in designing computer resource management systems is that computers are structured in multiple modular layers, such as hardware, operating system, and network. Each layer is complex and designed independently without full knowledge of the other layers. Therefore, computers must have modular resource controllers for each layer that are robust to modeling limitations and the uncertainty of inuence from other layers. Existing designs either rely heavily on ad hoc heuristics or lack modularity. We present a design with multiple Structured Singular Value (SSV) controllers from robust control theory for systematic and efcient computer management. On a challenging computer, we build a two-layer SSV control system that signicantly outperforms state-of-the-art heuristics.
more »
« less
- Award ID(s):
- 1763658
- PAR ID:
- 10095941
- Date Published:
- Journal Name:
- Proceedings of the IEEE Conference on Decision & Control
- ISSN:
- 2576-2370
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Fuzzy logic controllers can handle complex systems by incorporating expert’s knowledge in the absence of formal mathematical models. Further, fuzzy logic controllers can effectively capture and accommodate uncertainties that are inherent in real-world controlled systems. On the other hand, Robot Operating System (ROS) has been widely used for many robotic applications due to its modular structure and efficient message-passing mechanisms for the integration of system’s components. For this reason, Robot Operating System is an ideal tool for developing software stacks for robotic applications. This paper develops a generic and configurable Robot Operating System package for the implementation of fuzzy logic controllers, particularly type-1 and interval type-2, which are based on either Mamdani or Takagi-Sugeno-Kang fuzzy inference mechanisms. This is achieved by employing a systematic object-oriented approach using the Unified Model Language (UML) to implement the fuzzy inference system as a single class that is composed of fuzzifier, inference, and defuzzifier classes. The deployment of the developed Robot Operating System package is demonstrated by implementing an interval type-2 fuzzy logic control of an Unmanned Aerial Vehicle (UAV).more » « less
-
Workflow management systems (WMS) are widely used to describe and execute large computational or data intensive applications. However, when a large ensemble of workflows is run on a cluster, new resource management problems occur. Each WMS itself consumes otherwise unmanaged resources, such as the shared head node where the WMS coordinator runs, the shared filesystem where intermediate data is stored, and the shared batch queue itself. We introduce Mufasa, a meta-workflow management system, which is designed to control the concurrency of multiple workflows in an ensemble, by observing and controlling the resources required by each WMS. We show some initial results demonstrating that Mufasa correctly handles the overcommitment of different resource types by starting, pausing, and cancelling workflows with unexpected behavior.more » « less
-
Reasoning about storage systems is challenging because these systems make persistence guarantees even if the system crashes at any point. To achieve these crash-safety guarantees, storage systems include recovery procedures to restore the system to a consistent state after a crash. Moreover, large-scale systems are structured as multiple stacked layers and can require recovery at multiple layers of abstraction. Formal verification can ensure that crash-safety guarantees hold regardless of when the system crashes. To make verification tractable, large-scale systems should be verified in a modular fashion, layer-by-layer in the software stack. Layered recovery makes modularity challenging because the system can crash in the middle of a high-level recovery procedure and must start over from the low-level recovery procedure. We present Argosy, a framework for machine-checked proofs of storage systems that supports layered recovery implementations with modular proofs. The framework is based on combinators for transition relations that are inspired by Kleene algebra, which provides a convenient formalism for specifying and reasoning about crashes and recovery. On top of this framework, we implement Crash Hoare Logic (CHL), the program logic used by FSCQ. Using the logic, we have verified an example of layered recovery featuring a write-ahead log on top of a disk, which itself runs by replicating over two unreliable disks. The metatheory of the framework, the soundness of the program logic, and these examples are all verified in the Coq theorem prover.more » « less
-
This paper questions a fundamental assumption by a modern operating system (OS): it must run in the same computer it manages. We show that for many desirable OS functions, embedded systems often do not have the necessary resources. By carefully offloading some OS functions to another more resourceful computer, e.g., the cloud, one not only immediately overcomes the local resource limits but also opens the door for interesting optimizations because the remote computer becomes an advantageous point of aggregation and coordination. We discuss the challenges to offloading OS functions and their potential solutions. We also share some preliminary results of offloading system initialization logic and dynamic memory management from a microcontroller-based embedded system.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- The DOI is not currently available.
