Master key terminology in Computer Science and Information Technology (Terms 201- 250)

Series Part 5: Terms 201 – 250

The World-Class IT & CS Keyword Encyclopedia

Explore 50 new foundational, advanced, and emerging keywords covering MLOps, Distributed Databases, Network Protocols, Hardware Architecture, and Modern Security.

201. CAP Theorem

Distributed Systems

Definition: A fundamental principle in distributed data store design stating that a system can simultaneously guarantee at most two of three properties: Consistency, Availability, and Partition Tolerance.

Real-World Use Case: Choosing between Cassandra (AP system favoring Availability) and relational databases like PostgreSQL (CP system favoring Consistency) during network partitions.

202. MLOps (Machine Learning Operations)

DevOps & MLOps

Definition: A engineering discipline aimed at unifying machine learning system development (Dev) and machine learning system operation (Ops) to automate model deployment and monitoring.

Real-World Use Case: Automated pipeline tools like MLflow or Kubeflow retraining and redeploying recommendation models as new consumer data streams arrive.

203. QUIC Protocol

Networking & Security

Definition: A general-purpose transport layer network protocol designed by Google that uses UDP as its base, drastically reducing connection latency and improving web performance over TCP.

Real-World Use Case: HTTP/3 leveraging QUIC to eliminate head-of-line blocking and speed up mobile web browsing over unstable cellular networks.

204. Tensor Processing Unit (TPU)

Algorithms & Hardware

Definition: An application-specific integrated circuit (ASIC) custom-developed by Google to accelerate matrix math computations for deep neural network machine learning workloads.

Real-World Use Case: Accelerating model training on TensorFlow frameworks for massive artificial intelligence systems like Gemini and Search indexing.

205. Raft Consensus Algorithm

Distributed Systems

Definition: A consensus algorithm designed to be understandable and maintainable, managing replicated logs across distributed computer nodes to ensure state agreement.

Real-World Use Case: Distributed key-value stores like etcd (used by Kubernetes) electing leader nodes and keeping configurations synced.

206. GitOps

DevOps & MLOps

Definition: An operational framework that uses Git repositories as the single source of truth for declarative infrastructure and application deployments.

Real-World Use Case: Using ArgoCD or Flux to automatically sync live Kubernetes clusters whenever pull requests are merged into a Git repository.

207. eBPF (Extended Berkeley Packet Filter)

Networking & Security

Definition: A revolutionary technology that allows sandboxed programs to execute safely inside the Linux operating system kernel without changing kernel source code or loading modules.

Real-World Use Case: High-performance cloud observability and security tools like Cilium and Falco inspecting system calls in real-time.

208. Bloom Filter

Algorithms & Hardware

Definition: A space-efficient probabilistic data structure used to test whether an element is a member of a set, allowing false positives but zero false negatives.

Real-World Use Case: Web browsers checking malicious URLs or databases like RocksDB skipping disk lookups for non-existent key requests.

209. Eventual Consistency

Distributed Systems

Definition: A consistency model used in distributed computing where data updates will eventually propagate across all replicas given sufficient time, guaranteeing consistency over time.

Real-World Use Case: Social media follower counts updating slightly out of sync on different user screens across worldwide datacenters.

210. Feature Store

DevOps & MLOps

Definition: A centralized repository designed to store, manage, curate, and serve standardized machine learning features for offline training and online real-time inference.

Real-World Use Case: Platforms like Feast sharing pre-computed user click history features across multiple fraud detection and ad-ranking models.

211. Cross-Site Request Forgery (CSRF)

Networking & Security

Definition: A security vulnerability that forces an end user to execute unwanted, unauthorized actions on a web application in which they are currently authenticated.

Real-World Use Case: A malicious link sending an automatic HTTP form POST request to transfer funds from an logged-in banking user's account.

212. Non-Volatile Memory Express (NVMe)

Algorithms & Hardware

Definition: An open communications interface protocol specifically engineered to access high-speed solid-state drives (SSDs) over a computer's PCIe system bus.

Real-World Use Case: Modern SSD storage achieving bandwidth speeds exceeding 7,000 MB/s for intensive enterprise data analytics.

213. Distributed Hash Table (DHT)

Distributed Systems

Definition: A class of decentralized distributed systems that provide a lookup service similar to a hash table, storing key-value pairs across a peer-to-peer network.

Real-World Use Case: Peer-to-peer file distribution protocols like BitTorrent and IPFS locating files without reliance on central servers.

214. Canary Deployment

DevOps & MLOps

Definition: A deployment pattern that slowly rolls out application updates to a small sub-percentage of real users before releasing it to the entire infrastructure.

Real-World Use Case: Routing 5% of production traffic to a new software release to monitor log error rates before a 100% full rollout.

215. Web Application Firewall (WAF)

Networking & Security

Definition: A protective security barrier that monitors, filters, and blocks HTTP web traffic to and from an application, stopping attacks like SQL Injection and XSS.

Real-World Use Case: AWS WAF or Cloudflare blocking malicious HTTP requests containing illegal SQL commands before reaching web application servers.

216. MapReduce

Algorithms & Hardware

Definition: A programming model for processing and generating large datasets across distributed computer clusters using a Map step followed by a Reduce aggregation step.

Real-World Use Case: Apache Hadoop parsing petabytes of web log data across thousands of server nodes simultaneously.

217. Two-Phase Commit (2PC)

Distributed Systems

Definition: An atomic commit protocol that ensures all nodes in a distributed database system either commit or abort a transaction together through Prepare and Commit steps.

Real-World Use Case: Financial banking software ensuring a debit operation in Bank A and credit operation in Bank B commit atomically.

218. Blue-Green Deployment

DevOps & MLOps

Definition: A deployment technique maintaining two identical physical hardware environments (Blue for live production, Green for new build) to minimize application downtime.

Real-World Use Case: Switching load balancer traffic instantly from Blue to Green environment, enabling zero-downtime upgrades and quick rollbacks.

219. Mutual TLS (mTLS)

Networking & Security

Definition: A method for mutual authentication where both the client and server present cryptographic X.509 certificates to verify each other's identity before establishing a connection.

Real-World Use Case: Microservices within Zero Trust cloud architectures securely identifying and encrypting service-to-service communication.

220. Instruction Pipelining

Algorithms & Hardware

Definition: A hardware design technique in microprocessors where multiple instruction execution stages (Fetch, Decode, Execute) overlap to increase instruction throughput.

Real-World Use Case: Modern x86 and ARM CPUs executing multiple CPU instruction steps during single clock cycles.

221. Change Data Capture (CDC)

Distributed Systems

Definition: A set of software patterns used to detect and capture row-level data modification events in a database and stream them in real time to downstream systems.

Real-World Use Case: Debezium streaming PostgreSQL row updates into Apache Kafka to update Elasticsearch indices instantly.

222. Data Drift

DevOps & MLOps

Definition: The phenomenon where the statistical distributions of real-world input data change over time, causing machine learning model accuracy to decay.

Real-World Use Case: E-commerce shopper buying patterns changing during economic shifts, requiring automated model retraining triggers.

223. Software-Defined Networking (SDN)

Networking & Security

Definition: An approach to networking that separates the control plane (which decides how to route traffic) from the underlying data plane (which forwards the traffic).

Real-World Use Case: Cloud administrators dynamically reconfiguring network routes and firewall rules via software APIs without manually configuring physical switches.

224. LRU Cache (Least Recently Used)

Algorithms & Hardware

Definition: A memory caching algorithm that evicts the least recently accessed elements first when capacity limits are reached, maintaining items using HashMaps and Doubly Linked Lists.

Real-World Use Case: Operating system page replacement and web browser memory caches discarding old web assets.

225. Gossip Protocol

Distributed Systems

Definition: A decentralized communication protocol inspired by epidemic virus spreading, where computer nodes periodically share state metadata with randomly selected neighbors.

Real-World Use Case: Apache Cassandra clusters monitoring cluster node health membership without single points of failure.

226. Immutable Infrastructure

DevOps & MLOps

Definition: An infrastructure management approach where servers are never modified or patched in place after deployment; updates require destroying old instances and spinning up new ones.

Real-World Use Case: Deploying fresh Docker containers or AWS AMI images instead of manually SSHing into servers to install software updates.

227. OAuth 2.0

Networking & Security

Definition: An industry-standard open authorization framework enabling third-party applications to obtain limited access to user accounts without sharing passwords.

Real-World Use Case: "Sign in with Google" or "Sign in with GitHub" buttons allowing mobile apps access to basic profile info securely.

228. RISC-V

Algorithms & Hardware

Definition: An open-standard Reduced Instruction Set Computer (RISC) Instruction Set Architecture (ISA) that allows royalty-free hardware chip design.

Real-World Use Case: Custom IoT microcontrollers and AI silicon processors designed without paying proprietary licensing fees to ARM or x86 vendors.

229. Write-Ahead Logging (WAL)

Distributed Systems

Definition: A database reliability technique where changes are recorded sequentially in an append-only log on disk before being applied to actual database pages.

Real-World Use Case: Ensuring ACID database durability so transactions can be fully recovered even during sudden system power outages.

230. FinOps (Cloud Financial Operations)

DevOps & MLOps

Definition: An operational cultural discipline bringing financial accountability to the variable spend model of cloud computing, helping engineering teams optimize cloud costs.

Real-World Use Case: Tracking cloud resource tagging metrics to shut down unutilized cloud test servers automatically on weekends.

231. Zero-Knowledge Proof (ZKP)

Networking & Security

Definition: A cryptographic method enabling one party to prove to another party that a statement is true without revealing any information beyond the validity of the statement itself.

Real-World Use Case: Proving you are over 21 years old to a web app without disclosing your actual birthdate or full legal identity.

232. NUMA (Non-Uniform Memory Access)

Algorithms & Hardware

Definition: A multiprocessing computer memory design where memory access time depends on the memory location relative to a processor core.

Real-World Use Case: Enterprise multi-socket servers allocating process memory on RAM sticks attached directly to the local CPU socket.

233. Consistent Hashing

Distributed Systems

Definition: A distributed hashing scheme that operates such that when a hash table ring scales up or down, only K/n keys need to be re-mapped on average.

Real-World Use Case: Distributed caches like Memcached dynamic load balancing cache keys smoothly when nodes join or fail.

234. Continuous Integration (CI)

DevOps & MLOps

Definition: The software practice of automating the integration of code changes from multiple developers into a central repository followed by automated building and testing.

Real-World Use Case: GitHub Actions running unit tests, linters, and security scanners automatically whenever a software developer commits new code.

235. Server-Side Request Forgery (SSRF)

Networking & Security

Definition: A web security vulnerability that allows an attacker to induce the server-side application to make HTTP requests to an unintended internal resource.

Real-World Use Case: Forcing a web application to fetch cloud metadata IP addresses (e.g., 169.254.169.254) to leak AWS IAM security tokens.

236. Big O Notation

Algorithms & Hardware

Definition: A mathematical notation used in computer science to describe the worst-case limiting behavior and asymptotic complexity of an algorithm in terms of time or memory space.

Real-World Use Case: Comparing QuickSort O(N log N) time complexity against BubbleSort O(N²) time complexity when scaling data workloads.

237. Distributed Tracing

Distributed Systems

Definition: A monitoring method used to track and profile application requests as they travel across multi-tier microservice cloud architectures.

Real-World Use Case: Jaeger or OpenTelemetry passing unique Trace-IDs in HTTP headers to isolate slow database calls in microservice pipelines.

238. Continuous Deployment (CD)

DevOps & MLOps

Definition: The practice of automatically deploying every verified code commit passing CI test suites directly into live production environments without manual intervention.

Real-World Use Case: SaaS platforms pushing hundreds of small automated updates to production daily without manual approval delays.

239. JWT (JSON Web Token)

Networking & Security

Definition: An open standard (RFC 7519) defining a compact, self-contained method for securely transmitting claims between parties as a digitally signed JSON object.

Real-World Use Case: Authenticating stateless REST API requests by verifying JWT signatures sent in HTTP Bearer headers.

240. Cache Coherency

Algorithms & Hardware

Definition: The uniformity of shared resource data that ends up stored in multiple local L1/L2 caches in multi-core microprocessor systems (using protocols like MESI).

Real-World Use Case: Preventing CPU Core 1 from reading stale memory values when CPU Core 2 updates shared RAM locations.

241. Read-Through / Write-Through Cache

Distributed Systems

Definition: A caching architecture where the application treats the cache as the primary data store, relying on the cache layer to synchronously read from or write to the database behind it.

Real-World Use Case: Simplifying application code by delegating data fetching and persistence guarantees directly to Redis or Memcached enterprise layer plugins.

242. Model Drift

DevOps & MLOps

Definition: The degradation of a machine learning model's predictive capability due to fundamental changes in the relationship between input features and target outputs (concept drift).

Real-World Use Case: Housing price prediction models losing accuracy when regional economic policies alter real estate valuation logic.

243. Cross-Origin Resource Sharing (CORS)

Networking & Security

Definition: An HTTP-header based security mechanism that allows web servers to explicitly declare which foreign origins (domains) are permitted to load resources in web browsers.

Real-World Use Case: Configuring Access-Control-Allow-Origin HTTP headers so app.example.com can safely consume api.example.com backend services.

244. B-Tree / B+ Tree

Algorithms & Hardware

Definition: A self-balancing search tree data structure that maintains sorted data and allows logarithmic searches, sequential access, insertions, and deletions in disk storage systems.

Real-World Use Case: Powering indexing mechanisms in relational databases like MySQL (InnoDB) and file systems like NTFS/EXT4.

245. Dead Letter Queue (DLQ)

Distributed Systems

Definition: A specialized service queue holding messages that failed processing due to unhandled exceptions, malformed payloads, or network timeouts.

Real-World Use Case: RabbitMQ or AWS SQS automatically routing corrupt JSON payloads into a DLQ for developer debugging without blocking message streams.

246. Service Level Objective (SLO)

DevOps & MLOps

Definition: A target level of performance and availability service reliability (such as 99.9% uptime) agreed upon between service providers and customers.

Real-World Use Case: Engineering teams setting an SLO requiring 99% of API requests to respond in less than 200 milliseconds.

247. Content Security Policy (CSP)

Networking & Security

Definition: An added layer of web security that helps detect and mitigate Cross-Site Scripting (XSS) and data injection attacks by restricting approved content sources.

Real-World Use Case: Instructing browsers to execute JavaScript exclusively from trusted domains and forbidding inline script execution.

248. Branch Prediction

Algorithms & Hardware

Definition: A feature in modern microprocessor architectures that guesses which way a conditional instruction (if-else branch) will go before it is known for sure.

Real-World Use Case: Preventing CPU pipeline stalls by speculatively fetching and executing upcoming instructions in loop branches.

249. Read Replica

Distributed Systems

Definition: A read-only copy of a primary database instance used to offload read traffic queries and increase database read throughput across scaling clusters.

Real-World Use Case: Directing analytics dashboards and user web view requests to read replicas while routing user inserts exclusively to the primary writer DB.

250. Error Budget

DevOps & MLOps

Definition: The maximum allowable amount of system unreliability or downtime permitted under a Service Level Objective (SLO), balancing innovation speed with reliability.

Real-World Use Case: If a system's 99.9% uptime target leaves an error budget of 43 minutes per month, developers can push riskier features until that budget is exhausted.

🚀 Milestone Achieved: 250 CS & IT Keywords!

Congratulations on completing Part 5! You now have an authoritative 250-word Computer Science & IT dictionary covering full-stack systems, security, algorithms, and cloud infrastructure.

💬 Want Part 6 (Terms 251-300)? Leave a comment below with any specialized topics you want added!