Architecting for the cloud scability-availability
0 likes1,073 views
This document discusses architecting for scalability and availability in the cloud. It begins with an introduction to scalability, including definitions of scalability and why systems need to scale. It describes techniques for scaling such as load balancers, rule-based autoscaling, and scaling patterns like push and pull. It also covers CPU and I/O scaling. The document then discusses availability, including definitions and metrics. It describes different availability levels, calculating system availability, and how to define availability requirements based on understanding potential fault scenarios.
Architecting for the cloud scability-availability
- 1. Architecting for the Cloud Len and Matt Bass Scalability
- 2. Link to yesterday’s slides http://www.slideshare.net/lenbass/architecting- for-the-cloud-intro-virtualization-iaa-s
- 3. Outline • Introduction to scalability • CPU scaling • I/O scaling
- 4. Characteristic of cloud from NIST • On-demand self-service. A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service’s provider.
- 5. Scale in the Cloud • Many people think that you get scalability just by virtue of being in the cloud • This isn’t true • What the cloud gives you is the ability to quickly and easily add resources – It doesn’t guarantee that this results in additional capacity • Just like with security you need to design scalability in
- 6. What is Scalability? • (Problem definition) Scalability is the ability of a system to support growing amount of work. – May be from additional users – May be from additional requests from current users – May be from operational activities. • (Solution definition) Scalability is the ability to increase or decrease the resources available to your application by either changing the number of servers or disks or changing the size of the servers or disks.
- 7. Why scale? • Are more users always a good thing? – This is a cost/benefit question. – More users have benefits – presumably more people receive service and the organization more revenue. – More users have a cost – hardware, software, and personnel. • Do costs scale linearly with users? – For Netflix, the answer is yes. – For Linkedin, the answer is no.
- 8. The different aspects of scalability • Adding users – Large amounts of new users may require new computation facilities • Adding data – Large amounts of new data requires • More computation • Careful attention to the distribution of this data. • Adding computation – Computation is embedded in virtual machines – Elasticity means adding new virtual machines • Scaling should not impact existing activities • May need to scale by adding computation capacity (CPU) or by adding I/O capacity 8
- 9. Scaling Up vs Scaling Out • Scaling up means adding more capacity to existing hardware – More memory – More disk – Faster CPU or more cores • Scaling out means adding additional hardware – More systems
- 10. Costs in scaling out • Each virtual machine has a cost – per hour • Licensing costs. – Many software packages charge licenses per CPU or per (virtual) computer. – Every new instance that utilizes one of these packages incurs licensing costs • Personnel costs – In small to medium size organizations, one sysadmin can administer ~30 machines. – In large, highly automated organizations, one sysadmin can administer ~1000s of machines. – Movement called “DevOps” has as one goal the reduction of personnel costs in operations. (more on this later).
- 11. How much lead time for growth of number of users? • Some things are predictable – Seasonal variation. • Christmas • Tax season – Daily variation • Working hours or non-working hours in various time zones • Holidays – Promotions or special offers – Sporting events • Other things are not predictable – Being “SlashDotted” – News items – Rapid growth in popularity of a company. – Disaster
- 12. Managing growth in number of users • A lead time allows planning – Restructure database – Add or restructure software • When no lead time is available, elasticity of the cloud is the main mechanism.
- 13. Outline • Introduction to scalability • CPU scaling – Load balancers – Rule Based Scaling – Scaling Patterns • I/O scaling
- 14. Why have a load balancer? • Suppose there are too many users for a single instance of a service • The cloud allow us to create another instance of that service (elasticity) • We would like to have the half the users use one instance and half use the other • Two options: 1. Couple instances and users (half and half). This is accomplished by having users access an instance of a service directly by IP address. 2. Use an intermediary (load balancer) to distribute half of the requests to one instance and the other half to the other. Option 2 is preferable for a variety of reasons which we will see. 14
- 15. Load Balancing • Physically a load balancer is a box that looks like it belongs in a computer network.
- 16. Load Balancer Logically, a load balancer takes requests from clients and distributes them to copies of an application executing on multiple different servers Servers Clients Load Balancer
- 17. Message sequence – client makes a request Servers Clients Load Balancer
- 18. Message sequence- request arrives at load balancer Servers Clients Load Balancer
- 19. Message sequence – request is send to one server Servers Clients Load Balancer
- 20. Message sequence – reply goes directly back to client Servers Clients Load Balancer
- 21. Suppose Load Balancer Becomes Overloaded – Load Balance the Load Balancers
- 22. Hierarchy of Load Balancers • Server always sends message back to client. • Load balancers use variety of algorithms to choose instance for message – Round robin. Rotate requests evenly – Weighted round robin. Rotate requests according to some weighting. – Hashing – IP address of source to determine instance. Means that a request from a particular client always sent to same instance as long as it is still in service. • Note that these algorithms do not require knowledge of an instance’s load. That situation we will cover in a little bit.
- 23. Outline • Introduction to scalability • CPU scaling – Load balancers – Rule based scaling – Scaling Patterns • I/O scaling
- 25. Server • A server is a virtual machine without any software • A virtual machine can be allocated with varying amounts of memory, CPU, disk • Each variant has different cost, typically per hour
- 26. Machine Image • A machine image is a copy of the contents of the memory of a computer. • A machine image may be created from any contents of a computer. Some options: – Bare metal – With OS – With LAMP Stack • Linux • Apache HTTP Server • MySQL • PhP or Python • If licensed software is contained in the machine image, then a license fee is paid when it is loaded
- 27. Executable Virtual Machine • An executable virtual machine is created by loading a machine image into a server. • Executable virtual machine can then be – Booted – Paused – Shut down Machine Image Server
- 28. Adding/Removing Resources • Example shows two servers with one to be removed. • Could be N servers with one to be added or removed • Creating a new instance takes some time • Removing an instance also takes time – it must satisfy existing requests and be detached from existing connections.
- 29. Autoscaling group • An autoscaling group is a collection of instances that have been defined to be scaled together. • Typically these represent instances of the same application.
- 30. Creating an autoscaling group • An autoscaling group needs to know – Machine instance id – VM type – Scaling policy
- 31. Scaling Policy • Specify minimum, maximum, and desired number of instances • Can specify scaling based on time of day – E.g. scale up during 9:00-5:00 and down other times • Can scale based on average CPU usage – E.g. average CPU utilization <40% means delete instance – Average CPU utilization >60% means add instance. – Values come from monitor.
- 32. Outline • Introduction to scalability • CPU scaling – Load balancers – Rule Based Scaling – Scaling Patterns • I/O scaling
- 33. Scaling Patterns • Autoscaling implements Push Pattern for messages • Another pattern is Pull Pattern
- 34. Push Pattern
- 35. Push Pattern Description • Client sends a request (e.g. HTTP message) to the app in the cloud. • Request arrives at a load balancer • Load balancer forwards request to one of the VMs in the resource pool. • Load balancer uses scheduling strategy to decide which VM gets the request, e.g. dispatch to VM with lowest CPU utilization.
- 36. How does the load balancer know? • The load balancer knows CPU utilization of the VMs and it knows how many requests it (the load balancer) has received, and possibly how long it took to service the requests. It does not know application specifics such as how many requests a VM can process. • When resource pool is overloaded, new resources are allocated. • The monitor decides (based on controller rules) when new resources are needed. It must have direct insight into the VM instances in order to do this. Hence, the monitor utilizes a monitoring service provided by the cloud for each instance. 36
- 37. Pull architecture pattern (aka Producer- Consumer)
- 38. Pull architecture description • Each request from the client is application specific and typed. • The queue keeps separate queues for each application running on the VMs. • A VM requests the next message of a particular type (pull) and processes it. • The monitor can now see how long a request waits in a queue or the average queue length and this is an indication of the load on the VMs that have applications that service requests of that type.
- 39. Differences • Push is more responsive to requests. They are immediately forwarded to a service. There is a possibility that the service is overloaded. • Pull is less responsive since it relies on servers to de-queue messages. • In the pull architecture, a service polls for new messages even if there is nothing in its queue and this introduces overhead. • It is easier to monitor and control workload in the pull architecture since messages are application specific and typed.
- 40. Outline • Introduction to scalability • CPU scaling • I/O scaling – Multiple sites – Software techniques
- 41. I/O Scaling • Scaling out assumes scaling requirement is solved with more CPUs. • It may be that I/O is also a problem. – You may run your application in multiple sites – Half the clients go to one site, half to another
- 42. Questions when you have multiple sites How do clients know which site to use? How are databases used by the applications coordinated across sites (we defer this question).
- 43. Domain Name Server (DNS) Client sends URL to DNS DNS takes as input a URL and returns an IP address Client uses IP address to send message to load balancer for a site Site 1 Site 2 Domain Name Server Website.com 123.45.67.89 123.45.67.89 DNS
- 44. DNS with multiple sites • DNS server returns IP address of both sites. • DNS server will vary which address is listed first. • Client will, typically, choose first entry. Site 1 Site 2 Domain Name Server Website.com 123.45.67.89 456.77.88.99123.45.67.89 DNS
- 45. Outline • Introduction to scalability • CPU scaling • I/O scaling – Multiple sites – Software techniques
- 47. To Scale for I/O - Make the queue manager more sophisticated Key Value Store Publisher – takes values from key- value store and distributes them Clients
- 48. Summary • Scalability is the ability to respond to increasing or decreasing workload – Add CPU capacity through utilizing features of cloud provider – Add I/O capacity through • Distributing requests to multiple sites • Have fast message passing software
- 49. QUESTIONS?
- 50. Architecting for the Cloud Introduction to Availability
- 51. Outline • What is availability • Faults • Availability patterns
- 52. Outline • What is availability • Faults • Availability patterns
- 53. Cost of Downtime • According to a recent survey the average cost of unplanned downtime is $7,900/minute* • 91% of reporting companies have experienced an unplanned outage in the last 24 months • The average outage lasts 118 minutes • The average frequency of outages over a 24 month period were: – 10.16 limited outages – 5.88 local outages – 2.04 total outages * Emerson Network Power, Ponemon Institute Study 2013
- 54. Cost of Downtime II • As the previous numbers indicate downtime can be expensive • Experienced in August 2013 – New York Times had a 2 hour outage (stock price declined, twitter exploded, and Wall Street Journal dropped their fees to try and capture readership) – Google had between 1 – 5 minutes of downtime (~$500,000 direct loss and 40% reduction in overall web traffic) – Amazon had an outage of under an hour (> $5 million) • In addition to direct losses indirect losses are experienced – Loss of confidence, reputation, and good will – Productivity losses – Compliance penalties – …
- 55. Availability: a Business Concern • The availability of the business service impacts the earnings and associated value of an organization • If the organization relies on an IT system to deliver business service then the availability of the IT system impacts the value of the organization • In this section we are going to look at the availability of the system – We want to keep in mind, however, that the objective is the availability of the business service
- 56. What Is Availability? • Availability in general refers to the degree to which a system is in an operable state • This is typically articulated as the percentage of time the system is available (or we’d like to have the system available) e.g. 99.99% • There are many related terms e.g. – Availability – Fault-Tolerance – Reliability
- 57. How is Availability Measured? Availability is typically measured as: MTBF MTBF + MTTR MTBF = Mean Time Between Failures MTTR = Mean Time To Repair
- 58. 9s Availability Downtime per Year 90% (1-nine) 36.5 days/year 99% (2-nines) 3.65 days/year 99.9% (3-nines) 8.76 hours/year 99.99% (4-nines) 52 minutes/year 99.999% (5-nines) 5 minutes/year 99.9999% (6-nines) 31 seconds/year !
- 59. Calculating System Availability I • Each component = 99% (3.65 days a year) • The overall system, however, has an availability that is the product of each component’s availability – 99% X 99% = 98% (7.26 days a year) 99% 99%
- 60. Calculating System Availability • Each component = 99% (3.65 days a year) • The overall system in this case, however, is based on the likelihood that both components would fail at the same time 1 – ((100% - 99%) X (100% - 99%) )= 99.99% (3.65 hours a year!!) Redundant Elements 99% 99%
- 61. Availability Measures • A couple of things to keep in mind – These measures refer to the mean not the minimum time between failures – As the MTBF increases the impact of MTTR decreases – As the MTTR approaches 0 the overall availability approaches 1 • Historically these measures were developed for hardware components
- 62. Availability Requirements • MTBF can be measured for operational systems • How do you predict the MTBF for a system that is yet to be built, however? • Does it make sense to use the previously defined availability measure as a requirement? • If not, how should requirements be articulated?
- 63. Actionable Requirements • Remember that as a business the concern is that the services are available as needed • In order to determine the likely availability of a system (or design) you must – Understand the likelihood that various kinds of faults could occur – Understand the impact of these faults on overall system availability • You must therefore translate the desired business objective into a set of fault scenarios
- 64. End to End Availability • Engineers often think about availability of some portion of the system e.g. – Availability of the database or web server • Organizations, however, are concerned with end to end availability • When thinking about availability requirements you should think about the organizational perspective – Once you’ve done this you’ll then need to map this to the engineering perspective
- 65. Requirements Vary • We start with the desired requirements from a business perspective • We then look at the system context to determine what faults might disrupt the desired behavior – This is likely an iterative process • One thing to keep in mind is that different business contexts imply different requirements • Consider the needs of Discreet Manufacturing vs. Continuous Manufacturing • Discreet manufacturing is when you manufacture discreet products – e.g. an automobile assembly line • Continuous process automation is when you manufacture things like chemicals or concrete • How might the systems respond differently in the event of a fault?
- 66. Example Scenario If a processor in one of the servers fails during peak load, the system shall continue to operate without dropping any of the current tasks and without any noticeable delay
- 67. Relationship to Goals • How does this scenario relate to availability goals? – It does not in and of itself guarantee a particular level of availability • This in conjunction with scenarios for other faults that could impact a service do improve availability, however • In order to understand how to think about the design we need to: – Identify the activities that require availability – Identify the related faults – Identify the desired response if the fault occurs
- 68. Outline • What is availability • Faults • Availability patterns
- 69. Fault Characteristics • “Fail silent” vs. “fail operational” – Fail silent when a component fails it no longer operates – Fail operational a component continues to operate (although not correctly) when a fault is present • Transient vs. deterministic – Some faults will always occur in a consistent way – Others may come and go intermittently • Some will look similar to other faults e.g. – A hung process, a processor crash, and a network outage can all look the same
- 70. What’s the matter with this $#@!#% computer … A System Can Fail Silently … Let’s look at an example interaction Client Machine Network Server FileSystem Hmm … what’s the best vegetarian restaurant in Bogota?
- 71. Symptoms of Faults • From an end users perspective many faults exhibit themselves similarly • These faults could all look the same to an end user: – A hung process – A crashed processor – A network outage – An overloaded element
- 72. Or Fail Operational … Client Machine Network Server FileSystem Carnes de Res is the best vegetarian restaurant??? Hmm … what’s the best vegetarian restaurant in Bogota?
- 73. Fault Manifestation • These types of faults could occur in any of the elements of the system • Depending on where they occur different mitigation strategies might be appropriate • As a result you need to – Analyze your system and determine what faults might occur – Identify the desired response if they do occur • This is called a fault model
- 74. Fault Model • A fault model describes the system faults that could disrupt the critical functionality • The fault model is going to depend on both the critical functionality and the specific architecture of the system • Once the fault model is identified you’ll need to describe the desired response if the fault occurs
- 75. Cost of Availability • We’ve established that downtime can be expensive • It’s also the case that “uptime” can be expensive – Implementing a mechanism to be resilient to faults can be expensive • We want to understand the cost and benefit for proposed strategies and select the set that make sense from a business perspective • This means the initial requirements might change …
- 76. Example • We want “appropriate” availability • A study has been done for mobile carrier customers – This study has determined that customers will tolerate 2 dropped calls per 100 calls made – As soon as the system drops 3 calls per 100 they will start to change providers • What does this say about the “appropriate” availability of the system?
- 77. Outline • What is availability • Faults • Availability patterns
- 78. Elements of Availability • Fault detection – The system recognizes that a fault has occurred • Masking faults – The system is able to continue to operate despite the fault • Recover from the fault – The system is able to repair the faulty element of the system
- 79. Fault Detection • There are standard “tactics” that we can use for fault detection • They don’t detect the same types of faults, however • They also have different “costs” – This cost can be in terms of effort or overhead of one kind or another • We need to understand something about the kinds of faults we are trying to detect before we can select the appropriate tactic
- 80. Detecting Silent Faults • It’s much easier to detect elements that fail silently • Essentially we monitor the “liveness” of the element where the fault could exist • Example tactics are: – Exceptions – Heartbeat – Ping/echo
- 81. Exceptions • When an anomalous or exceptional event occurs it can be detected by exception handlers • When the exception is “caught” an alternate path of execution is triggered • The exception handling code can notify other portions of the system of the issue • Doesn’t impose significant overhead on the system
- 82. Heart Beat • A component emits a regular “heart beat” • Another element will listen for this • If this heart beat is not detected it is assumed that the component is no longer operational • Does add overhead to the system • Only an indication of the “liveness” of the component
- 83. Ping/Echo • Similar to heart beat except a “watchdog” sends a ping and listens for a response • If no response is heard it is assumed the component is not operational • Requires more coupling than heart beat • Increases network traffic • Again it’s only an indication of the liveness of the component
- 84. Failing Operational • If an element or system fails operational it’s more difficult to detect • You don’t just monitor if the system responds but also need to determine if the results are “correct” • Example tactics include: – Exceptions – Voting – Check sum
- 85. Voting • You compare the response of multiple elements performing the same operation • If the results of one of the elements doesn’t match the others you assume it’s faulty • Can detect erroneous output • Adds overhead (must wait for multiple responses and compare)
- 86. Check Sum • A mathematical calculation that’s applied to a piece of data to determine if it’s been altered • Does add some processing overhead to the system • Can detect data corruption
- 87. Tolerating Faults • In many cases you realize that faults will occur – Particularly in large distributed systems • You can’t tolerate outages every time one of the nodes experiences a fault • You therefore need to hide the fact that the system has a faulty component • This is called “fault masking” • Again the strategies associated with masking the fault are going to be dependent on the kind of fault being masked
- 88. Strategies For Fault Masking • Modular redundancy • Rollback – Restoring the system to a previously identified “safe state” • Roll forward – “skipping” an operation that is causing a problem • Retrying an operation • Shedding load • …
- 89. Modular Redundancy • Redundant systems have multiple replicated elements (copies) – Not to be confused with load balancing approaches – The thing to realize is that the state is replicated across the copies • There are multiple strategies for software replication – Cold standby – Warm standby – Hot standby
- 90. Redundancy: Cold Standby • There are non-operational copies available • State is stored (e.g. in logs) but is not loaded on the copies until they are needed • When a failure occurs the state is reconstructed and the replica is introduced • Reduces operational overhead associated with maintaining copies • Increases MTTR
- 91. Cold Standby
- 92. Redundancy: Warm Standby • In this configuration you have a primary replica that is actively processing requests • You have passive replicas that are not actively processing requests although they are online • State is periodically loaded into the backup replicas • As with cold standbys the processing overhead is reduced • The MTTR is dependent on the state checkpoints (typically less than with cold standbys)
- 93. Warm Standby
- 94. Redundancy: Hot Standby • All copies are processing requests • All of the duplicate responses will be suppressed • The copies need to be synchronized continuously – Thus the processing overhead is increased as the number of replicas increases • The MTTR is reduced to virtually zero, however, in the event that one of the replicas fail
- 95. Hot Standby
- 96. Considerations • State management – If there is state that is managed in the replicated elements you need to worry about synchronizing state • State can be pushed to other elements … – This impacts other concerns such as performance or security, however – Caching commonly accessed data is a typical strategy for dealing with performance concerns • Kinds of replicas • Frequency of check pointing
- 97. State Management
- 98. Roll Back • Roll back is when you undo a transaction • You need to manage state appropriately – You need to define an atomic set of actions • This could be taking complete snap shot of system state or just roll back of a transaction
- 99. Roll Forward • Roll forward essentially skips a task and then applies the changes involved in the transactions • The system will then be in the state consistent with the desired change
- 100. Retrying an Operation • This is as simple as it sounds • When a given operation fails you retry it • It can be used in conjunction with a detection mechanism like exceptions
- 101. Shedding Load • Sometimes issues occur due to an overload situation • This can lead to: – Timing errors – Buffer overflows – Memory consumption issues • Shedding less critical load can help alleviate the problem
- 102. Strategies For Fault Recovery • Reboot – This could be a partial (e.g. restarting an application or process) or total system reboot • Removal of faulty component • Restore component to a previously identified safe state • …
- 103. Reboot • Rebooting the system can often correct the issue • This can also be done as a preventative measure • It can be a complete or partial reboot • There is such a thing as a “micro reboot” that takes milliseconds
- 104. Component Removal • If you have a faulty component you can remove it from service • You might try other remedies such as restarting first
- 105. Checkpointing State • You can periodically take a snap shot of the system • If at some point you have an issue, you can restore the system to the previously defined state • The more frequently you take a snap shot of the state the smaller the loss but the more overhead
- 106. Availability in the Cloud • From a high level achieving availability in the cloud is the same process as elsewhere – It needs to be designed in • That means you need to understand the faults that could occur • You then need to apply the appropriate decisions to achieve the desired result
- 107. Fault Model • We will give specific faults that occur later in the course – This requires first a better understanding of the architecture of the cloud • At this point it’s useful to understand that the cloud is made up of faulty components – Failures happen on a regular basis • There are mechanisms built in to handle this, but – They aren’t always successful – They don’t deal with application specific concerns – Some things that might be a fault for your application isn’t considered a fault by the infrastucture
- 108. Summary • Availability measures are not adequate for design • You need to be able to translate availability goals into a set of actionable requirements that identify the possible faults and desired responses • The approaches should support the desired responses in the event that a fault occurs
- 109. Questions??