CS162 Operating Systems and Systems Programming Lecture 27 ManyCore OS and Peer-to-peer SystemsRequests for Final topicsSecurity TermsSecurity Problems: Buffer-overflow ConditionThe Morris Internet Worm: The beginning of chaosTiming Attacks: Tenex Password CheckingDefeating Password CheckingTypes of Parallel MachinesManyCore Chips: The future is on the wayBerkeley PARLabPar Lab Research OverviewPARLab approach to parallel programmingTraditional Parallel OSSome Tricky Things about Parallel OSsSlide 15ManyCore opportunities: Rethink the SinkSpatial PartitioningSpatial Partitioning and ApplicationsMinimalismUser-Level Protected MessagingFault Isolation and Optimistic ConcurrencyRealtime OS/Embedded ApplicationsMultiCore and RealtimeAdministriviaPeer-to-Peer: Fully equivalent componentsResearch Community View of Peer-to-PeerWhy the hype???OceanStoreUtility-based InfrastructureOceanStore: Everyone’s Data, One Big Utility “The data is just out there”Key Observation: Want Automatic MaintenanceExample: Secure Object StorageOceanStore AssumptionsPeer-to-Peer for Data LocationPeer-to-Peer in OceanStore: DOLR (Decentralized Object Location and Routing)Stability under extreme circumstancesPeek at OceanStore MechanismsOceanStore Data ModelSelf-Verifying ObjectsOceanStore API: Universal Conflict ResolutionTwo Types of OceanStore DataThe Path of an OceanStore UpdateSelf-Organizing Soft-State ReplicationArchival Dissemination of FragmentsAside: Why erasure coding? High Durability/overhead ratio!Extreme Durability?Differing Degrees of ResponsibilityClosing View on Peer-to-PeerPeer-to-peer Goal: Stable, large-scale systemsExploiting Numbers: Thermodynamic AnalogyExploiting Numbers: The Biological InspirationWhat does this really mean?Problems?ConclusionsCS162Operating Systems andSystems ProgrammingLecture 27ManyCore OS andPeer-to-peer SystemsDecember 10, 2007Prof. John Kubiatowiczhttp://inst.eecs.berkeley.edu/~cs162Lec 27.212/10/07 Kubiatowicz CS162 ©UCB Fall 2007Requests for Final topics•Some topics people requested:–Dragons: too big of a topic for today–ManyCore Systems–Parallel OSs –Embedded OSs–Peer-to-Peer Systems (OceanStore)–Virtual reality/enhancement–Quantum Computing•Today:–A couple of topics to finish from last time–ManyCore/Parallel OS–Embedded OS (realtime systems)–Peer-to-Peer Systems (OceanStore)•Other Topics:–Come look for me at office hours (Or any other time)Lec 27.312/10/07 Kubiatowicz CS162 ©UCB Fall 2007Security Terms •Virus:–A piece of code that attaches itself to a program or file so it can spread from one computer to another, leaving infections as it travels–Most attached to executable files, so don’t get activated until the file is actually executed–Once caught, can hide in boot tracks, other files, OS•Worm:–Similar to a virus, but capable of traveling on its own–Takes advantage of file or information transport features–Because it can replicate itself, your computer might send out hundreds or thousands of copies of itself•Trojan Horse:–Named after huge wooden horse in Greek mythology given as gift to enemy; contained army inside–At first glance appears to be useful software but does damage once installed or run on your computerLec 27.412/10/07 Kubiatowicz CS162 ©UCB Fall 2007Security Problems: Buffer-overflow Condition#define BUFFER SIZE 256int process(int argc, char *argv[]){char buffer[BUFFER SIZE];if (argc < 2)return -1;else {strcpy(buffer,argv[1]);return 0;}}Before attackAfter attack•Technique exploited by many network attacks–Anytime input comes from network request and is not checked for size–Allows execution of code with same privileges as running program – but happens without any action from user!•How to prevent?–Don’t code this way! (ok, wishful thinking)–New mode bits in Intel, Amd, and Sun processors»Put in page table; says “don’t execute code in this page”Lec 27.512/10/07 Kubiatowicz CS162 ©UCB Fall 2007The Morris Internet Worm: The beginning of chaos•Internet worm (Self-reproducing)–Author Robert Morris, a first-year Cornell grad student–Launched close of Workday on November 2, 1988–Within a few hours of release, it consumed resources to the point of bringing down infected machines•Techniques–Exploited UNIX networking features (remote access)–Bugs in finger (buffer overflow) and sendmail programs (debug mode allowed remote login)–Dictionary lookup-based password cracking–Grappling hook program uploaded main worm programLec 27.612/10/07 Kubiatowicz CS162 ©UCB Fall 2007Timing Attacks: Tenex Password Checking•Tenex – early 70’s, BBN–Most popular system at universities before UNIX–Thought to be very secure, gave “red team” all the source code and documentation (want code to be publicly available, as in UNIX)–In 48 hours, they figured out how to get every password in the system•Here’s the code for the password check:for (i = 0; i < 8; i++) if (userPasswd[i] != realPasswd[i]) go to error•How many combinations of passwords?–2568?–Wrong!Lec 27.712/10/07 Kubiatowicz CS162 ©UCB Fall 2007Defeating Password Checking•Tenex used VM, and it interacts badly with the above code–Key idea: force page faults at inopportune times to break passwords quickly•Arrange 1st char in string to be last char in pg, rest on next pg–Then arrange for pg with 1st char to be in memory, and rest to be on disk (e.g., ref lots of other pgs, then ref 1st page) a|aaaaaa | page in memory| page on disk •Time password check to determine if first character is correct!–If fast, 1st char is wrong–If slow, 1st char is right, pg fault, one of the others wrong–So try all first characters, until one is slow–Repeat with first two characters in memory, rest on disk •Only 256 * 8 attempts to crack passwords–Fix is easy, don’t stop until you look at all the charactersLec 27.812/10/07 Kubiatowicz CS162 ©UCB Fall 2007Types of Parallel Machines•Symmetric Multiprocessor–Multiple processors in box with shared memory communication–Current MultiCore chips like this–Every processor runs copy of OS•Non-uniform shared-memory with separate I/O through host –Multiple processors »Each with local memory»general scalable network –Extremely light “OS” on node provides simple services »Scheduling/synchronization–Network-accessible host for I/O•Cluster–Many independent machine connected with general network –Communication through
View Full Document
Unlocking...