Skip to main content

Medium network design issues and proxy servers

Short answer

Key design issues are scalability, performance, availability, security, manageability, cost, addressing, and wireless support, addressed with a hierarchical model, VLANs, redundancy, and QoS. Proxy servers are managed by placing them behind the firewall, configuring ACLs and authentication, enabling caching and content filtering, logging usage, controlling bandwidth, and providing high availability.

What are the major design issues with its design criteria to be considered for a medium size network? Also discuss how proxy servers are managed for this network.

A medium-size network serves roughly 100-1000 users in one campus or office with several departments, servers and an internet link. Its design must meet current needs and grow without a redesign.

Major design issues and criteria

IssueDesign criterion
ScalabilityHierarchical model (access, distribution, core); modular addressing; spare ports and rack space
PerformanceEnough bandwidth (Gigabit to desktops, 10G backbone); low oversubscription; QoS for voice and video
Availability / reliabilityRedundant links and core devices, RSTP or link aggregation, UPS, backup ISP
SecurityFirewall, DMZ for public servers, VLAN separation, ACLs, authentication, antivirus
ManageabilityStandard equipment, SNMP monitoring, labelled cabling, documentation, central logging
CostCost-effective equipment that meets need; consider total cost of ownership
Addressing and routingPrivate addressing with NAT, VLSM, DHCP, summarisation and a simple routing protocol (OSPF)
Flexibility / standardsOpen standards (IEEE 802.3, 802.1Q, TIA-568) so equipment from different vendors works
Wireless supportController-managed APs, separate staff and guest SSIDs

Managing proxy servers

A proxy server sits between the clients and the internet. Clients send web requests to it, and it fetches the pages on their behalf. Common software is Squid.

Placement. The proxy is placed in the server segment (or DMZ) behind the firewall. Clients are pointed to it by browser settings or the WPAD file, or traffic is redirected transparently by the router or firewall (port 80 to 3128).

Management tasks

  1. Access control: define ACLs by IP range, user, time and site category.
  2. Authentication: require login (LDAP or Active Directory) so usage can be tied to a user.
  3. Caching: tune cache size, refresh rules and object size to cut bandwidth use and speed up repeated access.
  4. Content filtering: block unwanted sites and file types, optionally with a filter list (for example SquidGuard).
  5. Logging and monitoring: review access logs and reports to find heavy users and problems.
  6. Bandwidth control: delay pools or QoS to limit large downloads.
  7. High availability: two proxies with load balancing or failover; for larger sites, a cache hierarchy (parent and sibling proxies).
  8. Maintenance: regular updates, log rotation, backup of configuration.
Sample Squid settings
http_port 3128
acl lan src 192.168.10.0/24
acl blocked dstdomain "/etc/squid/blocked.txt"
http_access deny blocked
http_access allow lan
http_access deny all
cache_mem 512 MB
 Clients --> [Proxy/Cache] --> [Firewall] --> Internet
              |  cache hit: reply locally
              |  log, ACL, auth

✓ Answer checked by Claude Opus 5.5 (high effort)

All Internet and Intranet chapter 5 questions →