CS5412: Spring 2012 Cloud Computing

CS5412: Spring 2012 Cloud Computing

CS5412 Spring 2014 (Cloud Computing: Birman) 1 CS5412: SPRING 2014 CLOUD COMPUTING Lecture 1 Ken Birman 2 Welcome to CS 5412... A course dedicated to the technology behind cloud computing! In my country of Khazackstan, many excellent hacker. We hack CS5412 Spring 2014 (Cloud Computing: Birman) cloud, steal private stuff of whole world! Cloud Computing: The Next New Thing 3

A general term for the style of computing that supports web services, search, social networking Increasingly powerful and universal Enables a new kind of massively scaled, elastic app Our goal: understand the technology of the cloud, its limitations, and how to push beyond them CS5412 Spring 2014 (Cloud Computing: Birman) 4 Todays Cloud: Surprisingly limited

Big data, updates by owner Dominated by reads Index... search... share Monetized by advertising, sales CS5412 Spring 2014 (Cloud Computing: Birman) Tomorrows cloud? 5 Big data, updates by owner Dominated by reads Index... search... share

Monetized by advertising, sales eHeal th CloudBa nk High assurance Real-time control Runs everything Monitized by roles eChauff er GridClo ud CS5412 Spring 2014 (Cloud Computing: Birman)

6 Clouds are hosted by data centers Huge data centers, far larger than past systems Very automated: far from where developers work. Often close to where power is generated (ship bits... not watts) Packed for high efficiency. Each machine hosts many applications (usually in lightweight virtual machines to provide isolation) CS5412 Spring 2014 (Cloud Computing: Birman) Scheduled to keep everything busy (but Clouds are cheaper and winning

7 Range in size from edge facilities to megascale. Incredible economies of scale Approximate costs for a Technology Cost in Cost in Cloud small-sized Large small size center (1K Advanta Data Data ge Center Center servers) and a larger, 50K Network per $13 per 7.1 server $95 center. Mbps/

Mbps/ Storage Administra tion month month $2.20 per GB/ month $0.40 per GB/ month 5.7 Each data center is 11.5 times the size of a football field ~140 >1000 7.1

CS5412 Spring 2014 (Cloud Computing: Birman) servers/ Servers/ Administra Administr Slide provided by Roger Barga, Head of Cloud Computing, Microsoft Key benefits? 8 Machines busier, earn more $s for each $ investment Applications far more standardized

Hardware handled a whole truckload at a time Automated management: few sys admins needed Power consumed near generator: less wastage Data center runs hot, wasting less on cooling Can rent resources rather than owning them Supports new, extremely large-scale services Elasticity to accomodate surging demands Can accumulate and access massive amounts of data But must read or process it in a massively parallel way Enables overnight emergence of major companies, but scalability model does require new programming styles, and imposes new limits CS5412 Spring 2014 (Cloud Computing: Birman) Assurance properties 9

Unfortunately, todays cloud Has a limited security model focused on credit card transactions Weakens consistency to achieve faster response times: the cloud is inconsistent by design Pushes many aspects of failure handling to clients Model supported by the CAP and FLP theorems, which are cited by many application designers CS5412 Spring 2014 (Cloud Computing: Birman)

Acronyms 10 CAP: A theorem that says one can have just two from {Consistency, Availability, Partition Tolerance} FLP: A theorem that says it is impossible to guarantee live fault-tolerance in asynchronous systems (here, live certain to make progress) BASE: A cloud computing methodology that seeks Basically available soft-state services with eventual consistency and is popular in the outer layers (first tier) of the cloud. The opposite of ACID ACID: A database methodology: offers guaranted {Atomicity, Consistency, Isolation and Durability}. CS5412 Spring 2014 (Cloud Computing: Birman) CS5412: How to do better! 11

Future cloud will need stronger guarantees than we see with todays cloud How can we achieve those? Are strong guarantees scalable? Betting that the cloud will win Cheaper than other options... ... and the cheaper option usually wins! But technology also advances over time, CS5412 Spring 2014 (Cloud Computing: Birman)

which helps! Making the cloud highly assured 12 Find ways to overcome limitations like FLP and CAP Define new assurance goals that might still be forms of security and consistency but are easier to achieve Only consider things that are real enough to be implemented and demonstrated to scale well and perform in a way that would compete with todays cloud platforms. A practical mindset. But use theoretical tools when theory helps with goals. CS5412 Spring 2014 (Cloud Computing: Birman) CS5412: Topics Covered

13 Well treat the cloud as having three main parts The client side: Everything on your device The Internet, as used by the cloud Data centers, which themselves have a tiered structure Like a dedicated and personal computer Yet massively scaled with many moving parts Special theme: CS5412 Spring 2014 (Cloud Computing: Birman) high assurance The Old World and the New

14 Old world: we replicated servers for speed and availability, but maintained consistency New world: scalability matters most of all Focus is on extremely rapid response times Amazon estimates that each millisecond of delay has a measurable impact on sales! But our premise is that we can have scalability and also have other guarantees that todays cloud lacks CS5412 Spring 2014 (Cloud Computing: Birman) High Assurance: Many (conflicting) goals 15

Security: Only correctly authorized users (who are properly authenticated) can perform actions Privacy: Data doesnt leak to intruders Rapid response despite failures or disruption Consistency and coordinated behavior Ability to overcome attacks or mishaps Guarantee that center operates at a high level of efficiency and in a highly automated manner Archival protection of important data CS5412 Spring 2014 (Cloud Computing: Birman) Must ask many questions 16

If we were to run high assurance solutions on todays cloud, what parts of the standards would limit or harm our assurance properties? Goal is to leverage the cloud or even run on standard clouds, yet to improve on normal options ThisCS5412 forces us to look hard at how things Spring 2014 (Cloud Computing: Birman) work Todays cloud focuses on easy stories 17

Which is better: Multithreaded servers? CS5412 Spring 2014 (Cloud Computing: Birman) Todays cloud focuses on easy stories 18 Which is better: Multithreaded servers? Or multiple singlethreaded servers? CS5412 Spring 2014 (Cloud Computing: Birman) Which scales best? 19 Build it the easy way!

One VM per server Server handles one user Make the server single threaded if possible Why? Better fit to the hardware (no lock/memory contention) Quicker way to build it, reuses existing stuff CS5412 Spring 2014 (Cloud Computing: Birman) 20 Some of todays rules of thumb Built from things that already exist and already work, as much as possible Expect that each 10x scaleup will still break

things and that much of your work will be on fixing them When feasible, go for no brainer scalability Armies of cheap machines and cheap storage A form of brute force solution Success stories of todays cloud often are applications that naturally fit this approach CS5412 Spring 2014 (Cloud Computing: Birman) Acronyms! (How to be a party bore) 21 One issue with the cloud is that it has a million acronyms: IaaS, SaaS, PaaS, SOAP, AWS, EC2, S3...

These make for a very confusing landscape! But a business perspective on the cloud only needs to focus on a few of them, as a starting point What does the aaS mean? Cloud vendors sell services aaS == as a Service ISO New England, Oct. 2013 The Important *aaS options 22 Infrastructure. (IaaS: Infrastructure as a Service) Cloud vendor rents you some hardware A network, perhaps a wide-area network A machine, always virtual but perhaps just for you A file server, again virtual, but you can save files in it

They operate this for you, and you pay for what you think you need (or sometimes, for what you use) And they sell backup services too For example, you could rent a private Internet from AT&T, or 500 computers from Amazon EC2 AWS is elastic: you rent and pay by the hour AWS can accommodate huge swings in your needs ISO New England, Oct. 2013 The Important *aaS options 23 Software. (SaaS: Software as a Service)

Cloud vendor runs some software that you use remotely Classic example: SalesForce.com has a sophisticated infrastructure that manages your sales contact data In effect you outsource your sales support system and SalesForce.com runs it for you Other SaaS options: accounting, billing, email, document handling, shared files They also apply patches, fix bugs ISO New England, Oct. 2013 The Important *aaS options 24 Platform. (PaaS: Platform as a Service)

Cloud vendor creates a sophisticated platform (typically a software environment for some style of computing, or for database applications) Your folks use it to create a custom solution Cloud vendor runs your solution in an elastic way They promise that if you use their PaaS solution, youll benefit from better scalability, performance, ease of development or other advantages ISO New England, Oct. 2013 The Important *aaS options 25 Platform. (PaaS: Platform as a Service) ISO New England, Oct. 2013

26 these arent the whole cloud The cloud mixes many models Some integrate humans into the loop, such as outsourced audio-to-text, or Amazons Mechanical Turk There are companies with specialized roles Akamai: The most famous data hosting company, especially successful for storing videos and images that are used in your web pages. They specialize in rapid data delivery DoubleClick: You leave a frame on your web page, they put the perfect advertisement for this particular user in it ISO New England, Oct. 2013

There are even cloud HPC systems! (Rent 27 But some standards pervade The cloud really took off as an outgrowth from web sites and browsers First we had browsers, HTML (a use of XML), HTTP, SSL Then people had the idea of doing client server computing using browser web pages! Called SOAP. A program makes a method call on a remote server they encode it as a special web page this is sent to the server just as if it was a web request from a browser (in fact you can do it by hand) result comes back in a special SOAP web page, extracted and returned to the calling program. Voila!

ISO New England, Oct. 2013 Web interoperability 28 In fact the web is about interoperability It is very easy to integrate Data from multiple sources (e.g. Netflix sends you a web page but in fact the video comes via Akamai) Different styles of computing (e.g. Weather.com fills a page with their content (the images come from Akamai), but the weather forecasts are from HPC computing systems and the advertisements are from DoubleClick. The ads might include a video hosted on YouTube, but Akamai might be the real source that sends the data By agreeing that at the end of the day, web pages are the lingua franca a great leap forward happened ISO New England, Oct. 2013

29 (Web pages are inefficient) The encodings used in the web are terribly inefficient, though So they made browsers extensible You get plug ins from Adobe, GZip, Microsoft, and those plug-ins extend the browser to understand special data representations Modern browsers can download and run full programs coded in Javascript, Silverlight, Caja or even true Java and these programs can do anything at all ISO New England, Oct. 2013 Open source 30

The cloud has hugely benefitted from open source (basically, source for programs is made available to customers), free open source (same, but no fee for use), and open development (many developers at many companies contribute). In fact nothing about the cloud demands open. But these are certainly powerful factors that help explain the vibrant cloud ecosystem. ISO New England, Oct. 2013 Open source debate 31

Many companies debate open source Quite a few have policies against it Yet they run Linux on their servers, build programs in C++ using gcc, allow employees to install their favorite browser add-ons, use Mono to create Linux versions of their Windows applications Java compiles to JIT code that reverse compiles back to Java source Believe me: You use open source even if you think you dont! (You probably even have employees who contribute to some open source projects) ISO New England, Oct. 2013

Deeper connection to cloud 32 The cloud is a world of open standards For the first time, the cloud tore down the high protectionist walls of proprietary products At many levels, we can see how things work and jump in and modify things Plug-and-play from the client system into the network and right up to the datacenter! The cloud is a world of easily interconnected component technologies that play together nicely And openness has been a key enabler in this happening ISO New England, Oct. 2013

33 So whats cloud computing? In some sense, the term means nothing! If you make full use of modern off-the-shelf computing products and systems, you are a cloud computing user You cant really buy non-cloud systems anymore The Internet and cloud standards are built into everything You can block some features, but it is surprisingly hard to create a cloud-free computing system (one of many reasons it is so easy to break into many systems) ISO New England, Oct. 2013 Can a cloud host high-assurance apps?

34 Motion sensor, fall-detector Healthcare provider monitors large numbers of remote patients Medication station tracks, dispenses pills Integrated glucose monitor and Insulin pump receives instructions wirelessly Cloud Cloud Infrastructure Home healthcare Infrastructure application Which matters more: fast response, or durability of the data being updated? 35

Mrs. Marsh has been dizzy. Her stomach is upset and she hasnt been eating well, yet her blood sugars are Lets stop the oral diabetes high. medication and increase her insulin, but well need to monitor closely for a week Cloud Cloud Patient Records Infrastructure Infrastructure DB Tradeoffs determine speed and What if we were doing online monitoring? 36

Update the monitoring and alarms criteria for Mrs. Marsh as A follows C Response delay seen by end-user would also include Internet latencies B D Soft-state first-tier Seservice Local response delay Confirmed

Execution timeline for an individual firsttier replica nd Se nd Se nd flu sh Durability matters more for patient records. But a monitoring system lives in the moment and mostly needs speed Which matters more: consistency or fast response? 37 Air Traffic Controllers depend on consistent data With a single

server ATC thisDBisnt hard to Safe for US guaranteeAir 221 to land? CS5412 Spring 2014 (Cloud Computing: Birman) Which matters more: consistency or fast response? 38 But suppose we replicate the server? Designate one as primary ATC DB Safe for US Air 221 to land? CS5412 Spring 2014 (Cloud Computing: Birman) Backup

Which matters more: consistency or fast response? 39 Failure detection will be key to consistency ATC Otherwise could endATC upDBwith two Safe for Air DB primaries! France 31 to Safe for US take off? CS5412 Spring 2014 (Cloud Computing: Birman) Air 221 to land? Cloud computing: A world of tradeoffs! 40

Cloud computing systems Overcome failure by replicating services But have no standard way to decide which server is in charge for a given service Easiest form of failure detection is by timeout But this might not be accurate: a network partitioning problem will look like a failure Maybe just some connections will fail And if the network then recovers, the old ATC service might not even know that we think it crashed! CS5412 Spring 2014 (Cloud Computing: Birman) Replication is central throughout

41 How to scale? Just add more replicas, balance load Fault-tolerance? If something crashes but has replicas, the impact is localized and other servers can take over Elasticity? Launch new replicas or shut some down What makes replication hard are cases where we need to think about coordination, concurrency control... If we dont worry about such things, may even be able to reuse existing applications! CS5412 Spring 2014 (Cloud Computing: Birman) Thrashing: Illustrates that 10x concern 42

With small-scale replication, IPMC is a big win But IPMC storms can occur in a data center with many replicas and heavy But it update rates worked in 12000 Wild10000 load swings, heavy loss rates, the lab! 8000 thrashing 6000 messages /s 4000 2000 0 250 400

550 time (s) 700 850 High assurance in the cloud 43 Todays cloud is built with simple components and yet even so, exhibits problems like split brain behavior, thrashing, rolling failures, other issues Companies spending a fortune to eliminate such issues They can limit scalability Tomorrows cloud thus poses a deep question

Will it be limited to simple applications? (Cloud Computing: Birman) OrCS5412 canSpring we2014 migrate application like health How will CS5412 approach such a complex set of problems? 44 Well take a step-by-step approach First look at properties of the client platform

Next consider Internet and its evolution under pressure of the cloud (e.g. for controlled routing, higher availability, better security) Finally focus on the data center and look at it tier by tier from the first tier inwards CS5412 Spring 2014 (Cloud Computing: Birman) At each level look at assurance issues 45 High assurance means different things in each layer A client depending on a browser worries about apps, personalization, connectivity, mobility, web-site spoofing, viruses, keystroke logging, privacy... The network worries about efficient routing, BGP problem, DDoS attacks, authenticating The cloud worries about maintaining rapid

response, balancing load, automating management, consistency, fault-handling, CS5412 Spring 2014 (Cloud Computing: Birman) etc. CS5412 Gets more technical as we go 46 For the first few weeks, well be more engineering oriented, because the first kinds of issues are ones that center on how scaled-out systems are built But then as we focus more on replicated processing and replicated data, well bring more theory into the picture Fault-tolerance will round off our investigation. Well explore many fault models but limit ourselves to ones seen in practice. We wont do as much on security. CS5412 Spring 2014 (Cloud Computing: Birman)

CS5412: Grades 47 Approximately 25 lectures, with [0-5] surprise quizzes (0-20% of your grade). Must be in class on time to take quizzes. No makeups! We maintain videonotes, in case you miss a lecture. Since some people will be ill or out of town, can miss a quiz without any negative impact on grade. Cloud projects (80%-100%), can be done on your own or in team. Deliverables every week or two. Course is curved to a B+/A-

CS5412 Spring 2014 (Cloud Computing: Birman) CS5412: Organization 48 Professor Birman gives most lectures Course roughly parallels his textbook Many assigned readings from textbook but they arent really required per-se; intended to help you understand the material Any quiz would focus on material covered in class because the goal of the quizzes is to ensure that you actually are coming to class We have two full-time TAs with office hours Web page has contact info and more details CS5412 Spring 2014 (Cloud Computing: Birman)

CS5412: Projects 49 Wide range of topics (well suggest many, or you can propose one of your own) Must meet with a TA twice during the semester to discuss topic, then report progress Graded by TA and Prof. Birman at end of semester Projects tackled by two people are expected to be more ambitious. Team CS5412 Spring 2014 (Cloud Computing: Birman) gets single grade 50

Examples of projects from 2012 Integrate Isis2 with Live Objects Build services of the kind Amazon uses for system monitoring using Code Partitioning Gossip Simulate and/or experiment on flow control for large scale replicated data sets, find best approach Implement a realistic Air Traffic Control system with high assurance properties (or a health care system) CS5412 Spring 2014 (Cloud Computing: Birman) Explore best options for wide area file 2014 Projects? 51

Long list on our course web page Some really good ones from the smart power grid, but flexibility to do something less ambitious or even to propose a project of your own Must reflect the themes and material of the course CS5412 Spring 2014 (Cloud Computing: Birman) CS5412: Textbook 52 Well be using Kens textbook Written as a teaching tool

Ken doesnt earn royalties on it! Available on reserve in library if you prefer not to own a copy Reading assignments will often be from the book but we may also assign a few published papers CS5412 Spring 2014 (Cloud Computing: Birman) Background assumed? 53 Solid understanding of computer archictectures, good programming skills including threads Some basic appreciation of how networks work, how operating systems

work, virtualization Prior exposure to distributed computing not required or expected CS5412 Spring 2014 (Cloud Computing: Birman)

Recently Viewed Presentations

  • Molecular Geometry and Bonding Theories

    Molecular Geometry and Bonding Theories

    VSEPR model assumes that the electron pairs around the central atom of a Lewis structure occupy space, whether they are bonding pairs or lone pairs, and the most stable arrangement of electron pairs (the one with the lowest energy) is...
  • Sensi FAQ - KeyPad Lockout - Mouser Electronics

    Sensi FAQ - KeyPad Lockout - Mouser Electronics

    All settings and schedule will remain unchanged. I don't want my employees to know the secret code to unlock the thermostat. ... iOS App version: 1.1.0 or greater. Android version: 1.091 or greater. I just installed my thermostat and it...
  • Sensation and Perception

    Sensation and Perception

    Afterimages Hearing Our auditory sense We hear sound WAVES The height of the wave gives us the amplitude of the sound. The frequency of the wave gives us the pitch if the sound. The Ear Transduction in the ear Sound...
  • Evidence & Policy How to influence policy using

    Evidence & Policy How to influence policy using

    Evidence and Analysis Estimate child poverty and deprivation Estimate child disparities by wealth, gender, regions, ethnic group Use DHS & others Analyze characteristics and causes of child poverty Qualitative methods /participatory app. ... non-conditional cash transfers Some examples cont ...
  • Skeletal System

    Skeletal System

    Vertebral prominens - useful landmark. Thoracic Vertebrae. Long spinous processes. Rib facets. Lumbar Vertebrae. Large bodies. Thick, short spinous processes. Sacrum. Five fused bertebrae. Median sacral crest. Dorsal sacral foramina. Posterior wall of pelvic cavity.
  • Ideas Why is innovation important? Innovation is key

    Ideas Why is innovation important? Innovation is key

    The North and West Yorkshire area has nine higher education institutions, more than any UK area outside London and representing one of the largest concentrations of HEIs in Europe. Together, these HEIs have around 130,000 students and produce 40,000 graduates...
  • Bacterial Damage to Host Cells - University of Southern ...

    Bacterial Damage to Host Cells - University of Southern ...

    Parenteral route Bite, puncture, injection, wound Most microbes have a preferred portal of entry Streptococci when inhaled may cause pneumonia; when ingested they do not A few microbes cause illness no matter how they enter May cause different illness based...
  • cal-span.org

    cal-span.org

    Riparian HabitatFunctions, Values and Beneficial Use Protection. Extended Seasonal Flows. Sediment Trapping. Pollutant Filtering . Wildlife Habitat. Rain Drop Dissipation. Surface