Download Comprehensive Guide to Workload Distribution, Load Balancing, and Cloud Architecture UPDA and more Exams Medicine in PDF only on Docsity!
Comprehensive Guide to Workload Distribution, Load
Balancing, and Cloud Architecture UPDATED ACTUAL
Questions and CORRECT Aswers
Workload Distribution architecture: how can the workload be evenly distributed among IT resource By horizontally scaling and a load balancer. Resulting workload distribution architecture educes both IT resource over utilization and under utilization to an extent dependent upon the sophistication of the load balancing algorithms and runtime logic what is Workload Distribution Architecture is a design pattern focused on horizontally scaling IT resources to prevent bottlenecks, ensure high availability, and optimize performance C- Context C represents the real-time state of the system and the rules of the environment. C-context dynamic part of architecture include? Resource Availability, Cost/Quota, Locality/Latency, Priority/SLA, Affinities Essential components of architecture The Load Balancer, Resource Cluster, Cloud Usage Monitor, Resource Replication
what is the core architecture component of a WDA? the load balancer load balancer also known as? traffic controller /front door. sitting between the user and the pool of resources. what specific entity that executes Func F? Load Balancer is the specific entity that executes the mapping function F what are the 3 critical sub functions that load balancer is responsible of WDA? Request Routing: It receives incoming traffic and determines which server in the resource set (R) is best suited to handle it.
- Health Checking: It continuously pings the resources. If a server stops responding, the Load Balancer automatically removes it from the set R so that no users experience a "404 Not Found" or a timeout.
- Session Persistence (Sticky Sessions): what is WDA efficiency dependent on? on the logic to move workload. aka Distributed strategies (Algorithms) Static algorithms used in WDA (^) Round Robin: Simply goes down a list. Request 1 goes to Server A, Request 2 to Server B, and so on.
- Weighted Round Robin: If Server A is twice as powerful as Server B, the Load Balancer will send two requests to A for every one sent to B. Dynamic algorithms used in WDA? Least Connections: The Load Balancer checks which server currently has the fewest active users and sends the new request there.
- Least Response Time: It pings the servers; the one that responds fastest gets the next task. how does work load distribution happen in cloud ecosystem? multiple depths within the cloud ecosystem: L4, L7, Geographic layer what are the benefits of resource pooling architecture design pattern? many physical or virtual resources be treated as a single shared capacity that can be allocated dynamically. what is resource pooling architecture based off? on the use of one or more resource pools, in which identical IT resources are grouped and maintained by a system that automatically ensures that they remain synchronized what can also be apart of resource pooling architecture? Pay-Per-Use Monitor
Workload Distribution (The Delivery) his is the top layer. Once the resources are pooled and scaled, this architecture ensures that the incoming traffic is spread evenly across them so no single node becomes a bottleneck. what is the functional flow? Step A (Resource Pooling), Step B (Dynamic Scalability), Step C (Workload Distribution) what happens if the architecture doesn't have distribution? You can scale up to 100 servers, but all the traffic will still hit the first server and crash it while the other 99 sit idle what does Elastic resource capacity architecture design pattern focuses on? focused on the dynamic adjustment of a system's total computing power to match fluctuating demand. ERCA ensure what in the equation F: WxC-> R that R is not a fixed size but a stretchable entity. ERCA coordinates with? operates through a coordination between the Cloud Usage Monitor, a Scaling Policy, and the VIM (Virtual Infrastructure Manager) what is an automation Engine? is the "execution core" of a cloud or infrastructure system. It is the software component responsible for coordinating, managing, and executing automated tasks based on predefined rules or templates. what additional mechanisms can be included in the cloud architecture? cloud usage monitor. Elasticity vs scalability Scalability is the capability of a system to handle more work (the "potential") (ability to grow).
- Elasticity is the automated action of matching resources to that work in real- time(ability to grow and shrink). Service Balancing Architecture a specialized design pattern focused on distributing requests specifically at the service or application layer (Layer 7 of the OSI model) Difference between Load balancer and service LBs look at IP address meanwhile SBs understands the nature of the service being balancing? requested (e.g., an API call vs. a video stream) and routes traffic to the most appropriate serviceinstance based on that context
how come replication alone doesn't help unless incoming In distributed systems, services are typically replicated to achieve: request are distributed effectively across replicas? • higher throughput
- lower latency
- fault tolerance Why Service Load Balancing Exists? replication alone does not help unless incoming requests are distributed effectively across replicas.
- Service load balancing exists to answer:
- How can client requests be distributed across multiple service instances such that performance, availability, and correctness goals are met under changing load and failures? How does service load balancing work? functions as a sophisticated "Switchboard." It doesn't just send a user to a random server; it examines the request content and the service health to make a decision. what is the flow when a user sends a request? Request Inspection: A client sends a request (e.g., api.csun.edu/v1/research).
- Service Discovery: The Service Balancer looks up a "Service Registry" to see which specific containers or VMs are currently running the /research service.
- Contextual Routing: It checks the status of those instances (load, latency, or version).
- Delivery: 4 pillars of SBA? Service Broker/Registry -
- It maintains a live list of all active service instances and their current IP addresses.
- Service Usage Monitor -
- It tracks metrics specific to the service, such as request-per-second (RPS) or error rates
- Service Discovery Engine -
- It automatically finds new service instances as they are spun up by the Automation Engine.
- Resource Cluster Differences from Workload Distribution and service Workload Distribution: Balances traffic across hardware/nodes. (e.g., "Send this balancing? user to Server A because Server B is full.") Service Balancing: Balances traffic across logical services. (e.g., "Send this user to the 'Search' service cluster because their request starts with /search.") why is Load balancing algorithms significantly more Because the Service "intelligent than simple network level distribution? Balancer can see the actual content of the request, it can use sophisticated logic to decide which specific service instance should handle the work. how are the SB algorithms divided into? divided into Static (fixed logic) and Dynamic (real-time state-aware) categories.
