Looking Glass: Test Your Network Connection Before You Commit

Ever rented a server only to find out the connection to your region is terrible? That's exactly why Looking Glass exists. Think of it as a test drive for data center networks - you get to check speed, routing, and stability before making any commitments.

Looking Glass is basically a diagnostic tool that lets you peek into a data center's network infrastructure from the inside. Instead of guessing whether a server location will work for your needs, you can run real tests and see actual performance metrics. It's like having a network engineer on staff, except it's free and available 24/7.

What You Can Actually Do With It

The tool covers four main testing scenarios. First, you can check basic availability and response times to any IP address or domain you specify. Second, you can trace the exact path your data packets take from the data center to your destination, including every router and network hop along the way. Third, you can diagnose connection quality issues like packet loss or unstable routing. And fourth, you can verify DNS records and domain configurations.

For businesses running latency-sensitive applications or managing international traffic, these diagnostics matter a lot. A few extra milliseconds can mean the difference between smooth video streaming and constant buffering, or between responsive gaming and lag-induced frustration.

Running Your First Speed Test

When you land on a Looking Glass page, you'll typically see a speed test option right at the top. This basic test measures download and upload speeds between your current location and the data center, plus it shows your ping time. Many providers also offer test files in different sizes - usually 100MB, 1GB, or 10GB options - so you can simulate real-world data transfers.

The speed test gives you a quick snapshot, but the real power comes from the network diagnostic commands. Each serves a specific purpose and reveals different aspects of connection quality.

Understanding Ping Results

Ping is the simplest diagnostic tool. You enter an IP address or domain name, and the system sends small data packets to that destination, measuring how long each round trip takes. The output shows you the average response time in milliseconds and whether any packets got lost along the way.

Good ping times depend on distance. Within the same continent, you'd expect 10-50ms. Cross-continental connections typically range from 100-200ms. Anything consistently above 300ms will feel noticeably slow for interactive applications. More concerning than high latency is packet loss - even 1-2% packet loss can cause noticeable problems for VoIP calls or online gaming.

👉 Get consistently low-latency connections with strategically located data centers worldwide

Tracing the Route Your Data Takes

Traceroute shows you every single hop your data makes between the data center and your target destination. Each line in the output represents one router or network node, showing its IP address, hostname, and response time.

When you run traceroute, you might see entries like this: hop 1 might be the data center's gateway router responding in under 1ms, hop 5 could be a major ISP node in another city responding in 15ms, and hop 12 might be your destination server responding in 45ms total. This progression helps you pinpoint exactly where delays occur.

Sometimes you'll see asterisks instead of response times. That doesn't necessarily mean there's a problem - some routers are configured to not respond to traceroute probes for security reasons. What matters is whether your packets are reaching the final destination successfully.

The geographic path matters too. If you're connecting from Tokyo to a server in Amsterdam, but traceroute shows your packets routing through Los Angeles first, that's adding unnecessary latency. Efficient routing should take the most direct path possible.

MTR: The Upgraded Diagnostic Tool

MTR combines ping and traceroute into one continuous test. Instead of just showing you the route once, it keeps sending packets and collecting statistics over time. This reveals intermittent issues that single-run tests might miss.

The MTR output includes several columns worth understanding. The Loss% column shows packet loss at each hop - this is critical because even 5% loss at any point can degrade your connection quality significantly. The Avg column shows average latency, while Best and Worst show the range of response times. Wide gaps between Best and Worst indicate unstable routing or congestion.

For example, if you see 0% packet loss on the first 10 hops but suddenly 15% loss at hop 11, that specific router is likely experiencing problems. You can then research whether that node belongs to your ISP, a transit provider, or the destination network - which helps determine who to contact about the issue.

DNS Queries With Dig

The Dig command queries DNS servers to show you exactly how domain names resolve to IP addresses. This is useful when troubleshooting website accessibility issues or verifying that DNS changes have propagated correctly.

When you query a domain, the ANSWER SECTION shows you the DNS records. An A record maps a domain to an IPv4 address, AAAA records handle IPv6 addresses, MX records specify mail servers, and CNAME records create domain aliases. The TTL value tells you how long other DNS servers should cache this information before checking again.

If you're migrating a website to a new server, you can use Dig to confirm that the domain now points to the correct new IP address. If users in different regions are reporting they can't reach your site, Dig helps verify whether DNS is resolving correctly from various locations.

Practical Applications

Looking Glass tools become especially valuable in a few specific scenarios. Before committing to a dedicated server or colocation space, you can test the actual network quality from that data center to your target audience. If users report connection problems, you can run diagnostics to determine whether the issue is with your server, the data center's network, or somewhere in between.

For content delivery and streaming services, consistent low latency matters more than raw bandwidth. You might have a 10Gbps connection that looks great on paper, but if the routing is inefficient or certain hops experience regular packet loss, your users will have a bad experience. Running MTR tests during peak and off-peak hours reveals whether the network maintains quality under load.

👉 Test network performance to your specific region before choosing a server location

Reading Between the Numbers

Raw numbers only tell part of the story. A ping time of 150ms isn't inherently good or bad - it depends on the physical distance and what you're using the connection for. File transfers and web hosting can handle higher latency just fine, while real-time voice communication or gaming need sub-100ms response times.

Consistency matters as much as speed. A connection that averages 50ms but spikes to 200ms randomly will feel worse than one that stays steady at 75ms. Check the standard deviation in MTR results - high values indicate unstable connections even if the average looks acceptable.

Geographic routing logic also plays a role. Sometimes the shortest physical distance doesn't equal the fastest network path, especially when dealing with undersea cables or international borders. A server in Singapore might actually deliver better performance to Australia than one in Sydney, if the Singaporean data center has better peering arrangements.

When to Run These Tests

The best practice is testing during the hours when your users are most active. Network performance can vary significantly between 3 AM and 3 PM as congestion levels change. Run tests from multiple locations if your user base is geographically distributed - a data center that works great for European users might struggle to serve Asian traffic efficiently.

Also test during the trial period after ordering any hosting service. Most providers offer at least a few days to evaluate whether their infrastructure meets your needs. Use that window to run comprehensive diagnostics rather than discovering problems after you've already migrated production systems.

Looking Glass diagnostics give you the data to make informed decisions about server placement and network providers. The difference between guessing and knowing can save significant time and money, especially when scaling operations or serving latency-sensitive applications.