VXLAN explained: extending your L2 networks beyond the datacenter
VLANs cap at 4096 and don't cross routing. VXLAN encapsulates L2 in IP to stretch a network across racks, rooms and sites. Here's how.
Read the article →The blog
Network resources & analysis
Clear guides on IP transit, security, BGP and running an operator network.
Antivirus, EDR, XDR: properly protecting your endpoints and servers
Classic antivirus no longer cuts it against ransomware. EDR, MDR, XDR: we break down these endpoint protections and how to choose the right one.
Read the article →Colocation: how to choose the right datacenter (Tier, redundancy, PUE)
Not all datacenters are equal. Tier, N+1 redundancy, physical security, connectivity: here are the concrete criteria to host your infrastructure without surprises.
Read the article →24/7 NOC: why outsource your network monitoring
Monitoring a network around the clock in-house is expensive and needs rare skills. Here's what an outsourced NOC brings and how it concretely improves your availability.
Read the article →Connecting your sites: VPN, MPLS or SD-WAN — how to choose
Linking several corporate sites is done today in three main ways: VPN, MPLS and SD-WAN. We compare performance, security, cost and operations to help you choose.
Read the article →IP transit, peering and IXPs: how an operator carries your traffic
Transit, peering, IXPs: three complementary ways to connect a network to the Internet. Here's how they work and what they change for your performance.
Read the article →DDoS attacks: understanding L3, L4 and L7 threats — and how to stop them
Volumetric, protocol, application-layer: not all DDoS attacks are alike. A breakdown of the L3/L4/L7 layers and the defences that actually work.
Read the article →BGP made simple: announcing your prefixes and controlling your routing
BGP is the protocol that holds the Internet together. What it does, why you might need it, and what 'announcing your prefixes' really means.
Read the article →Moving to IPv6: why migrate now and how to get started
IPv4 is exhausted and expensive. IPv6 is no longer a distant option: here's why to migrate now and how to deploy dual-stack with confidence.
Read the article →RPKI and routing security: putting an end to BGP hijacks
BGP was built on trust, and that shows up as route leaks and hijacks. RPKI, ROAs and MANRS finally add guardrails. Here's how it works.
Read the article →Anycast explained: one IP, servers everywhere
Anycast lets you announce the same IP from many places worldwide: each user reaches the nearest site. A pillar of DNS, CDNs and DDoS defence.
Read the article →Reducing network latency: where it comes from and how to control it
Latency isn't about bandwidth. Physical distance, BGP hops, congestion, queuing: here's where it comes from and how a good network reduces it.
Read the article →Choosing an IP transit provider: the criteria that really matter
Price per Mbps tells you almost nothing about transit quality. Real capacity, routes, redundancy, SLA, DDoS, support: here's how to compare seriously.
Read the article →GRE tunnels: getting transit without a physical presence
No presence in the operator's datacenter? A GRE tunnel lets you establish a BGP session and benefit from transit and DDoS protection remotely, over your existing access.
Read the article →Blackholing and RTBH: the last-resort weapon against DDoS
When a volumetric attack saturates everything, blackholing (RTBH) cuts off traffic to the targeted IP outright. A radical countermeasure to use with judgement.
Read the article →BGP multihoming: staying online when an operator goes down
Depending on a single operator means going down when it goes down. BGP multihoming spreads that risk across several providers. Here's how to design it.
Read the article →