Date

Presenter

Topic

Feb 10

Daniel Fryer

Recon: Verifying File System Consistency at Runtime

File system bugs that corrupt file system metadata on disk are insidious. Existing file-system reliability methods, such as checksums, redundancy, or transactional updates, merely ensure that the corruption is reliably preserved. The typical workarounds, based on using backups or repairing the file system, are painfully slow. Worse, the recovery is performed long after the original error occurred and thus may result in further corruption and data loss.

We present a system called Recon that protects file system metadata from buggy file system operations. Our approach leverages modern file systems that provide crash consistency using transactional updates. We define declarative statements called consistency invariants for a file system. These invariants must be satisfied by each transaction being committed to disk to preserve file system integrity. Recon checks these invariants at commit, thereby minimizing the damage caused by buggy file systems.

The major challenges to this approach are specifying invariants and interpreting file system behavior correctly without relying on the file system code. Recon provides a framework for file-system specific metadata interpretation and invariant checking. We show the feasibility of interpreting metadata and writing consistency invariants for the Linux ext3 file system using this framework. Recon can detect random as well as targeted file-system corruption at runtime as effectively as the offline e2fsck file-system checker, with low overhead.

Feb 17

Ioan Stefanovici

Cosmic Rays Don't Strike Twice: Understanding the Nature of DRAM Errors and the Implications for System Design

Main memory is one of the leading hardware causes for machine crashes in today's datacenters. Designing, evaluating and modeling systems that are resilient against memory errors requires a good understanding of the underlying characteristics of errors in DRAM in the field. While there have recently been a few first studies on DRAM errors in production systems, these have been too limited in either the size of the data set or the granularity of the data to conclusively answer many of the open questions on DRAM errors. Such questions include, for example, the prevalence of soft errors compared to hard errors, or the analysis of typical patterns of hard errors.

In this project, we study data on DRAM errors collected on a diverse range of production systems in total covering nearly 300 terabyte-years of main memory. As a first contribution, we provide a detailed analytical study of DRAM error characteristics, including both hard and soft errors. We find that a large fraction of DRAM errors in the field can be attributed to hard errors and we provide a detailed analytical study of their characteristics. As a second contribution, we use the results from the measurement study to identify a number of promising directions for designing more resilient systems and evaluate the potential of different protection mechanisms in light of realistic error patterns. One of our findings is that simple page retirement policies might be able to mask a large number of DRAM errors in production systems, while sacrificing only a negligible fraction of the total DRAM in the system.

Feb 24

Phillipa Gill

Understanding Network Failures in Data Centers: Measurement, Analysis, and Implications

We present the first large-scale analysis of failures in a data center network. Through our analysis, we seek to answer several fundamental questions: which devices/links are most unreliable, what causes failures, how do failures impact network traffic and how effective is network redundancy? We answer these questions using multiple data sources commonly collected by network operators. The key findings of our study are that (1) data center networks show high reliability, (2) commodity switches such as ToRs and AggS are highly reliable, (3) load balancers dominate in terms of failure occurrences with many short-lived software related faults, (4) failures have potential to cause loss of many small packets such as keep alive messages and ACKs, and (5) network redundancy is only 40% effective in reducing the median impact of failure.

Mar 2

Peter Feiner

Comprehensive Kernel Instrumentation via Dynamic Binary Translation

Dynamic binary translation (DBT) is a powerful technique that enables fine-grained monitoring and manipulation of an existing program binary. At the user level, it has been employed extensively to develop various analysis, bug-finding, and security tools. Such tools are currently not available for operating system (OS) binaries since no comprehensive DBT framework exists for the OS kernel. To address this problem, we have developed a DBT framework that runs as a Linux kernel module, based on the user-level DynamoRIO framework. Our approach is unique in that it controls all kernel execution, including interrupt and exception handlers and device drivers, enabling comprehensive instrumentation of the OS without imposing any overhead on user-level code. In this paper, we discuss the key challenges in designing and building an in-kernel DBT framework and how the design differs from user-space. We use our framework to build several sample instrumentations, including simple instruction counting as well as an implementation of shadow memory for the kernel. Using the shadow memory, we build a kernel stack overflow protection tool and a memory addressability checking tool. Qualitatively, the system is fast enough and stable enough to run the normal desktop workload of one of the authors for several weeks.

Mar 23

Beom Heyn (Ben) Kim

Unity-VM: a Single system image for an individual PC user

    Today, PC users are having difficult time to manage their devices, since there are usually more than one or two PCs per user unlike a few years ago. This is because as the technology evolves the price of PCs drops quickly while the capability of them are significantly enlarged. Although users can enjoy the convenience of having multiple devices for different purposes – desktop PC for performance intensive tasks, laptops for versatility and mobile devices for mobile computing, this brings the management problems of each device which has different and separately managed system image including OS, Apps and all user data as well as personal settings.
    To deal with this problem, we are trying to implement the Unity-VM which is the tool providing a single system image across a user’s PC devices by migrating a VM containing the whole personal computing environment. By migrating the entire system image contained in VM along with a user, the user can work with each of device through a single computing environment and only have to manage one.
    There have been previous works like ISR and The Collective using the VM migration technology to solve the similar problem. Yet, they utilized either the central repository to put the suspended image of the machine or the portable storage devices like USB keys. Most of all, they can't handle distributed VM image. We believe this limitation is due to the lack of synchronization mechanism for the VM image including CPU state, memory state and disk state. With this, currently existing solutions can not support instant device switch for users. Also, they are relying on the reliability of the storage medium they are using as the central repository of the VM image.
    To resolve disk consistency issue and support the instant device switching, Unity-VM keeps track of the location of latest blocks distributed over multiple devices and support on-demand fetching of the proper version of data objects. Unity-VM utilizes the centralized meta-data server, the directory server, for keeping the location information. Also, we will support automatic replication for each page of the VM images over the multiple nodes to improve the reliability.

Mar 30

Nilton Bila

Jettison: Efficient Idle Desktop Consolidation with Partial VM Migration

Idle desktop systems are frequently left powered, often because of applications that maintain network presence or to enable potential remote access. Unfortunately, an idle PC consumes up to 60% of its peak power. Solutions have been proposed that perform consolidation of idle desktop virtual machines. However, desktop VMs are often large requiring gigabytes of memory. Consolidating such VMs, creates bulk network transfers lasting in the order of minutes, and utilizes server memory inefficiently. When multiple VMs migrate simultaneously, each VM's experienced migration latency grows, and this limits the use of VM consolidation to environments in which only a few daily migrations are expected for each VM. This paper introduces Partial VM Migration, a technique that transparently migrates only the working set of an idle VM. Jettison, our partial VM migration prototype, can deliver 85% to 104% of the energy savings of full VM migration, while using less than 10% as much network resources, and providing migration latencies that are two to three orders of magnitude smaller.

Apr 20

Billy Yi Fan Zhou

Information Leak Prevention for Android

Mobile devices play an increasingly important role in people’s lives. They are entrusted with large amounts of private data, but their mobility also makes them prone to loss and theft. The danger is even greater for mobile devices used in the corporate or government environment where they hold valuable data and are a target for espionage. Our research seeks to move the sensitive user data outside mobile device in order to prevent information leakage during loss or theft. Our first approach involves running Android on a secure server while streaming the screen to a mobile device using a custom VNC protocol. Our second approach involves partitioning Android applications such that the user interface portion of the application runs on the mobile device while portions of the application containing sensitive information runs on a server environment.