Comprehensive Overview of Server Architecture, Classification, and Key Components
This article provides a detailed, English-language overview of server hardware and architecture, covering non‑x86 and x86 classifications, instruction set families, measurement units, firmware components, memory and storage distinctions, cache hierarchy, CPU affinity, networking standards, and management protocols such as SNMP.
Servers are familiar to anyone working in IT, yet many details of server technology remain unclear; this article aims to give readers a thorough understanding of server fundamentals, starting with architecture and classification.
Based on system architecture, servers are divided into non‑x86 (including mainframes, minicomputers, and UNIX servers that use RISC or EPIC processors and specialized operating systems) and x86 servers (CISC architecture using Intel‑compatible CPUs and typically Windows OS).
The article explains instruction set families: CISC (Complex Instruction Set Computing), RISC (Reduced Instruction Set Computing), and EPIC (Explicitly Parallel Instruction Computing), noting that classification standards vary.
Key measurement units are described, such as rack unit (U) for height, capacity units for storage, and rate units (bit/s, B/s) for data transfer, as well as floating‑point operation metrics (FLOPS) for performance.
Essential server firmware components are outlined: BIOS, UEFI, CMOS, and BMC, each playing a role in hardware initialization, configuration, and management.
The distinction between memory (RAM) and storage (disk) is clarified with an office‑desk analogy, and memory frequency is explained as an indicator of speed measured in MHz.
Cache hierarchy (L1, L2, L3) and its function in bridging CPU speed and memory latency are detailed, including instruction vs. data caches and the role of higher‑level caches.
CPU affinity (processor affinity) is introduced as a technique for binding virtual CPUs or threads to specific physical cores to improve cache utilization and scheduling efficiency.
Networking concepts such as Ethernet auto‑negotiation, switch types (access, aggregation, core), stacking vs. cascading, and the differences between switching and routing are presented.
Storage networking topics include FC SAN zoning (hard vs. soft zones) and the TPC benchmark standards (TPC‑C, TPC‑D, etc.) for measuring transaction processing performance.
Management protocols are covered, focusing on SNMP (versions 1‑3) and its components: network management system (NMS), managed devices, and agents.
Architects' Tech Alliance
Sharing project experiences, insights into cutting-edge architectures, focusing on cloud computing, microservices, big data, hyper-convergence, storage, data protection, artificial intelligence, industry practices and solutions.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.