Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitchesiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
The only useful Solana RPC latency test runs from the same region as your application or users. Send the same JSON-RPC method, parameters, and commitment to each endpoint, repeat enough times to capture normal variation, and compare median and tail times with errors, timeouts, and returned-slot freshness. A quick response from a node that is behind the cluster can still produce stale application data.
What you are actually measuring
An RPC latency result is the elapsed wall-clock time between your client sending an HTTP request and receiving the response. It is not a pure measure of geographic distance. The observed interval can include DNS lookup and connection setup, network transit, provider routing, queueing, TLS, RPC method execution, and response transfer. A different method or request size can therefore produce a different result even when the endpoint and region do not change.
For a useful comparison, report the distribution of repeated requests rather than one unusually fast response. The median describes a typical request; p95 and p99 show the slower tail that users encounter during congestion or transient routing changes. Record successful responses, HTTP or JSON-RPC errors, connection failures, and timeouts alongside those timings.
Free tools Windows power users keep installed
One-click scans. No signup required.
Choose the correct test location and endpoint
Run from the deployment region
Use the server, container, CI runner, or cloud region that will make the production requests. If your users are distributed, run the same test from each important region and keep the results separate. A laptop test can follow a completely different route and should not be used to select an endpoint for a server elsewhere.
#1 Best Overall
- Powerful Turbo Fan:CAGIWIRU Electric Air Duster have reached speed up to 130000RPM,Can efficient remove dust and debris.With LED light and Three Adjustable Speed to meet with different cleaning tasks.
- With Attached USB C Fasting Charging cable.Charing 3 hours,Enjoy up to 240 minutes of use on the lowest setting, For third speed can use for 25mins.with four charging options to meet with your needs.
- Wide Application: attached 5 different nozzles,making it suitable for a variety of scenes, such as car,pet, pc, keyboards, and other electronic devices. It also serves for office and home clean duster.
- Handy Design & Portable:Cordless air duster electric nozzle is Light weight and small size, design and comfortable to hold and space saving, convenient to carry around. You can put it in your room or inside the car without taking up space.
- Economical &Reusable: Built durable materials with a long-lasting battery,our CAGIWIRU air duster it's a sustainable alternative to disposable air cans.
Document the vantage point for every result: cloud provider or network, region, IPv4 or IPv6 path if relevant, test date, and whether a proxy or VPN was involved. Anycast and provider routing can change the path without a change to your code.
Use the exact Solana cluster
Test the cluster your application will read or write. Mainnet, Devnet, and Testnet are separate networks, and an endpoint for one cannot stand in for another. Solana publishes public HTTP RPC endpoints for these clusters. The public Mainnet endpoint used in the example below is https://api.mainnet.solana.com.
Public endpoints are shared and rate-limited development infrastructure. A result from the public endpoint does not establish the capacity, consistency, or support you will receive from a private production service. Candidate private endpoints must be tested from your own deployment region under a representative request rate.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Keep every comparison variable fixed
Change only the endpoint when comparing providers. Keep these values identical:
- Region, host network, and client software.
- HTTP method, JSON-RPC method, parameters, and request body size.
- Commitment level:
processed,confirmed, orfinalized. - Number of requests, interval between requests, timeout, and test duration.
- Whether the connection is reused or newly opened for each request.
Commitment changes what state a node is allowed to return. processed is the newest state but can be rolled back; confirmed has been voted on by a supermajority; finalized provides the strongest confirmation. A test that uses different commitments is measuring different freshness guarantees as well as different response times.
Step-by-step HTTP latency test
1. Start with a lightweight health read
getSlot is a small read that makes it convenient to compare basic request round trips. Send JSON-RPC 2.0 over an HTTP POST with a JSON content type:
curl -sS -o /dev/null -w 'http_code=%{http_code} total=%{time_total}sn'
--connect-timeout 5 --max-time 15
-H 'Content-Type: application/json'
-d '{"jsonrpc":"2.0","id":1,"method":"getSlot","params":[]}'
'https://api.mainnet.solana.com'time_total is the complete request round trip observed by curl. It includes provider routing, queueing, and method execution; it does not isolate those components. Keep the HTTP status in the output so a fast error is not mistaken for a successful low-latency result.
2. Repeat instead of trusting one request
Run a controlled loop from the target region. The following example writes one elapsed time per line and marks failures as FAIL. Replace the URL with each candidate endpoint, and use the same loop for all candidates:
#!/usr/bin/env bash
set -u
url='https://api.mainnet.solana.com'
for i in $(seq 1 100); do
result=$(curl -sS -o /dev/null
--connect-timeout 5 --max-time 15
-w '%{http_code} %{time_total}'
-H 'Content-Type: application/json'
-d '{"jsonrpc":"2.0","id":1,"method":"getSlot","params":[]}'
"$url" 2>/dev/null) || result='000 FAIL'
printf '%s %sn' "$i" "$result"
sleep 0.2
doneOne hundred samples are a practical starting point, not a universal requirement. Run long enough to include the normal operating period that matters to you, and repeat during representative peak load if the application has a known busy window. Do not hammer a shared endpoint; respect its published rate limits and stop when errors begin to reflect test traffic rather than normal use.
Rank #2
- Screen Cleaner is Your Screen’s New Bestfriend! - EVEO is proud to present a screen cleaner that’s perfectly suited for your TV. Meet your new screen cleaner, computer screen cleaner, laptop cleaner, iPad cleaner, Macbook,tv cleaner or any other electronic device. Our screen cleaner spray and wipe along with the included plush microfiber cloth, easily removes dust, fingerprints & other smudges on the screen’s surface. Clean effectively for a streak-free screen
- Gentle on your Screens - Our screen cleaner spray and wipes is compatible to be used even for the most sensitive LCD, LED, CRT, and OLED screens in proper use. Be at ease knowing that this product is compatible with all the major brands: LG, SONY, Samsung, Sharp, iPad, iPhone, Android screens, Kindle, PC monitors - we’ve got you covered! Can be used as LCD screen cleaner, tv cleaner, smart tv screen cleaner, and more
- Premium Super Soft Plush Microfiber Cloth - Included in this efficient screen cleaning kit. This screen cleaner spray and wipe has our signature plush microfiber cloth specially designed to comply with screens and monitors and leave them streak-free. Has been shown to effectively absorbs stubborn stains, zero streaks, and no fiber shedding. Always use this together with our screen cleaner for the best results
- Easy-to-Use Screen Cleaner Kit + microfiber cleaning cloth - The EVEO Screen Cleaning Kit will easily be your new favorite tool when it comes to taking care of the things you love most, quick and easy solution for a streak-free screen. Simply screw on the sprayer gun on the screen cleaner bottle. Spray the cleaning liquid onto the microfiber cloth, then wipe the screen until desired cleanliness & shine is achieved.
- Quality Screen Care Kit Worth Purchasing - Enjoy a convenient Screen Cleaner anytime, anywhere you need it. Your satisfaction is our top priority. We stand behind the quality of our products and that’s why, you shouldn’t be worrying at all. We manufacture the Best Screen Cleaner. and can be used for laptop cleaning kit, laptop screen cleaner, tv screen cleaner for smart tv. MacBook screen cleaner, monitor cleaner or screen cleaner spray for electronic devices with microfiber cleaning cloth
3. Test a representative read
After the lightweight check, test the method your application actually depends on. getLatestBlockhash is a useful transaction-flow read and is included in the Solana Foundation monitor’s checks. Account reads, program-account queries, and larger responses can have different execution and transfer costs, so include those when they represent real traffic. Keep parameters and commitment identical across endpoints.
For example, a blockhash request with an explicit commitment can be sent as:
Recommended Free Tools
curl -sS -w 'nhttp_code=%{http_code} total=%{time_total}sn'
-H 'Content-Type: application/json'
-d '{"jsonrpc":"2.0","id":1,"method":"getLatestBlockhash","params":[{"commitment":"confirmed"}]}'
'https://api.mainnet.solana.com'Save the response body for this test. A 200 response can still contain a JSON-RPC error, and the returned value is needed for the freshness check described below.
Summarize the measurements correctly
For every endpoint, retain the individual samples and calculate:
- Success rate: successful JSON-RPC responses divided by attempts.
- Error rate: HTTP errors, JSON-RPC errors, connection failures, and malformed responses.
- Timeout rate: requests exceeding the same client timeout.
- Median: the 50th percentile of successful elapsed times.
- p95 and p99: the 95th and 99th percentiles, which expose tail behavior hidden by an average.
Do not silently discard failures when calculating reliability. You can report latency percentiles for successful requests and list failure rates separately; combining a timeout with a numeric latency would hide the operational impact.
If your log contains one seconds value per line, this small Python example calculates common percentiles. It deliberately leaves failed requests out of the latency distribution while counting them separately in your log review:
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
from pathlib import Path
values = []
failures = 0
for line in Path("times.txt").read_text().splitlines():
value = line.strip()
if not value or value == "FAIL":
failures += 1
continue
try:
values.append(float(value))
except ValueError:
failures += 1
if values:
values.sort()
def percentile(p):
index = (len(values) - 1) * p
low = int(index)
high = min(low + 1, len(values) - 1)
return values[low] + (values[high] - values[low]) * (index - low)
print({"count": len(values), "median_s": percentile(.50),
"p95_s": percentile(.95), "p99_s": percentile(.99),
"failures": failures})
else:
print({"count": 0, "failures": failures})Percentile conventions differ slightly between tools, so use the same calculation for every endpoint and record the method with your report. Solana’s guidance recommends comparing distributions, errors, and timeouts but does not define a location-independent latency threshold. There is no single millisecond value that makes an endpoint universally “fast.”
Check data freshness separately from speed
A node can answer quickly while lagging behind the cluster. For each successful read, compare the returned slot with a second healthy endpoint or another trusted reference tested at approximately the same time. Track the slot difference and how long a candidate remains behind, not just whether it answered.
Use the same commitment when making the comparison. A lower commitment can legitimately show a newer slot than a finalized request, so label the commitment next to every freshness measurement. If a provider is consistently behind, its low round-trip time does not make it suitable for state-sensitive operations.
Rank #3
- 【Large size】：Measuring 800 x 300 mm/ 31. 5×12 inches, which is wider and taller than others. Large mouse pad can perfectly fit your keyboard and mouse of any size, and there is also ample space for peripherals such as phones, tablets, etc.
- 【LED backlight mouse pad】：Blade Hawks RGB mouse pad provides 7 static modes and 7 dynamic modes will give you the options you are looking for. Power-off memory function remembers the last lighting mode you selected. The super glow fiber Chroma will give you a colorful gaming setup that you want.
- 【Anti-slip rubber base】：Made from Natural rubber, this gaming mousepad will firmly grip your desktop, allowing you to enjoy the ultimate precision in the games you are most passionate about. It also comes with a texture that enhances this ability even further.
- 【Smooth and waterproof surface】: Micro-textured cloth surface, for extremely precise mouse control and a smooth movement. When liquid splashes on the gaming mouse pad, it forms water droplets and slides down. Will not delay your work or gameplay.
- 【Plug and play】: This mouse and keyboard pad is powered by USB, plug and play, no driver required. One button control light modes. Press one time to change the lighting mode, press twice to change the brightness, press and hold about 3 seconds turn ON/OFF.
For an application that submits transactions, measure freshness in the workflow that users experience: submit the transaction, then measure the time until the signature status is observable at the required commitment. This end-to-end value includes transaction forwarding, validator processing, and confirmation; it cannot be inferred from a standalone getSlot request.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Measure WebSocket subscriptions as a different workload
HTTP timings do not describe subscription quality. If the application uses Solana WebSockets, open a real subscription from the target region and record:
- Connection establishment time and handshake failures.
- Time from a known on-chain event to notification receipt.
- Disconnect frequency and reconnect duration.
- Whether notifications missed during a disconnect are recovered after reconnecting.
Test recovery deliberately by interrupting the connection or allowing a controlled network failure. A provider that has excellent HTTP p95 latency can still lose events or take too long to restore a subscription.
Compare endpoints with a consistent scorecard
| Dimension | What to record | Why it changes the decision |
|---|---|---|
| Region and route | Actual client region, network, address family, proxy/VPN, and test period | Distance, peering, anycast, and routing determine which path was measured. |
| Latency distribution | Sample count, median, p95, and p99 for each method | Tail delays affect interactive requests even when the median looks good. |
| Reliability | Successes, HTTP/JSON-RPC errors, connection failures, and timeouts | A low median is not useful if requests frequently fail or exceed the client deadline. |
| Freshness | Returned slot, comparison slot, commitment, and observed lag | Fast responses from stale nodes can produce incorrect application state. |
| Workload | Methods, parameters, response sizes, and request rate | Heavy account reads and transaction-related calls can stress an endpoint differently from getSlot. |
| Connection behavior | WebSocket handshake, notification delay, reconnect, and recovery | Subscription applications need guarantees that HTTP tests cannot provide. |
| Production fit | Capacity, rate limits, support, scaling behavior, and fallback design | Development endpoints and isolated spot checks do not establish production readiness. |
Publish the test window and workload with the table. Provider routing and public endpoint limits are volatile, so an old result should not be treated as a permanent ranking.
Interpret differences without overclaiming
When one endpoint has a lower median
Confirm that its p95 and p99 are also acceptable and that its error, timeout, and freshness results are comparable. A lower median with a long tail may feel slower than a slightly higher but stable median in an interactive application.
When results vary widely between runs
Check whether the test changed region, IP family, DNS result, connection reuse, request rate, or commitment. Run the candidates in alternating order rather than testing one for an hour and the other later; this reduces the chance that time-of-day or congestion explains the entire difference.
When a provider is fast but behind
Treat freshness as a blocking issue for features that require current slots, balances, blockhashes, or transaction status. Investigate another endpoint or a provider configuration with a healthier node instead of optimizing the already-low request time.
When a private endpoint is slower than a public one
Verify that the private endpoint is in the intended region and that you are not comparing a warm, connection-reused public request with a new TLS connection each time. Then test under the expected concurrent load. Production capacity and support can matter more than a single idle-latency sample.
Troubleshooting a failed or misleading test
- DNS or connection failures: resolve the hostname from the deployment host, test IPv4 and IPv6 paths separately if your stack supports both, and check firewall egress rules.
- Frequent HTTP 429 or provider errors: reduce test rate, honor the endpoint’s limits, and repeat with a controlled workload. Do not interpret rate limiting as network latency.
- Requests exceed the timeout: retain them as failures, then try a longer diagnostic timeout only to distinguish a slow response from a dead connection. Keep the production timeout unchanged in the final comparison.
- Fast HTTP status but JSON-RPC error: parse the response body and classify JSON-RPC errors separately from transport success.
- Unexpectedly stale slots: compare the same commitment against a second endpoint and check whether the test method itself is reading a cached or lower-freshness result.
- Inconsistent measurements: pin the test host, method, parameters, request rate, and connection policy; capture timestamps so routing changes are visible.
Turn the test into a production decision
Select an endpoint only after it meets the application’s actual requirements for tail latency, reliability, freshness, and workload capacity from each important region. Define those requirements internally—for example, a maximum acceptable p99 and timeout rate for an interactive request—because Solana does not publish one universal latency target.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Rank #4
- 【Softer and More Comfortable】Vaydeer wrist rest has unique diamond pattern, which is the combination of softness and aesthetics. The materials of wrist rest are improved into higher quality memory foam and covered with silky smooth lycra. The computer wrist rest makes you as comfortable and cushiony as like rest your wrists on clouds.
- 【Ergonomic Wrist Saver】The wrist rests for keyboard and mouse comes with a 17.32×3.15×0.83 inch keyboard wrist pad and a 5.94×3.15×0.83 inch mouse wrist support. Based on ergonomic design, the unique concave shape is the perfect fit for your wrist joints. The wrist rest pad fits most computer keyboards and laptops, improve hand and wrist posture, release your wrist and arm stress.
- 【Non-Slip Rubber Bottom】Featuring an anti-skid silicone base on the bottom, this wrist keyboard support stays firmly in place on your desk, preventing the padding from sliding around, ensuring stable and consistent wrist support during extended computer sessions.
- 【Better Experience & Pain Relief】Our keyboard arm rest is beneficial to alleviate the soreness caused by direct contact and friction between your arm and a hard desk surface, reducing the risk of wrist fatigue or carpal tunnel. The soft texture of memory foam can evenly distribute the pressure around your wrists and provide good support with just enough give.
- 【Helpful in Multiple Scenarios】Whether you're working, studying, writing, typing, gaming, this keyboard and mouse rest combo is an essential accessory to add comfort and support to your hands and wrists. It’s also a great gift for men, women, family, friend, coworker, gamer, teacher, etc.
Use separate endpoints or a health-aware fallback when a single failure would interrupt the application. A fallback should be checked for the same freshness and commitment behavior; failing over to a fast but lagging node merely changes the failure mode. Monitor the same metrics continuously after deployment, because routes, cluster conditions, provider capacity, and public limits change over time.
For production traffic, assess private RPC access with capacity and support matched to your request volume. Treat the curl example as a repeatable measurement technique, not as proof that any endpoint is the fastest everywhere.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Build a repeatable regional test plan
When an application serves more than one geography, create one test record per region instead of averaging all locations together. Label each record with the cluster, endpoint, method, commitment, sample count, start and end times, and whether connections were reused. Running candidates in alternating order helps distinguish provider behavior from a temporary route or congestion event. Repeat the same plan at a quiet period and during the busiest period that matters to the application.
Keep raw responses for representative calls, not only summary numbers. They let you verify that every endpoint returned the same JSON-RPC result shape and that a nominal HTTP 200 was not hiding an application-level error. For freshness, retain the observed slot and comparison slot with timestamps so a later review can tell whether a lag was brief or persistent.
Use the resulting table as an engineering decision record rather than a permanent leaderboard. Public limits, anycast routes, private-provider capacity, and cluster conditions can all change. A scheduled rerun from the deployment regions catches those changes before they become user-visible.
Questions readers commonly ask
Is ping enough to measure Solana RPC latency?
No. Ping measures an ICMP path, while an RPC call includes HTTP connection behavior, provider routing, queueing, JSON-RPC execution, and response transfer. Measure the actual POST request and the method your application uses.
Which Solana RPC endpoint is fastest from my location?
There is no universal answer. The fastest endpoint depends on your region, route, request method, commitment, load, and time of day. Test each candidate from the machine or region that will send production traffic, then compare median, p95, p99, failures, and freshness.
Should I test Mainnet, Devnet, or Testnet?
Test the cluster your application will use. Mainnet, Devnet, and Testnet are separate networks, so a result from one cluster does not establish performance on another.
How many requests should a latency test send?
One hundred requests is a practical starting point for a basic comparison, but it is not a universal requirement. Use enough samples to cover the operating period that matters, repeat under representative peak load, and respect shared-endpoint rate limits.
Best Value
- 【No Drill Mounted】Cable management tray can be assembled in as little as 3 minutes or less-simply. Side clips easily clampe the edge of desk, pass through the holes on the sides, and secure excess cables with baskets. Desk cable management supports inward or outward installation to meet your needs in different scenarios, and provides you with great convenience when collecting and organizing wires.
- 【Rubber Pad Protects Desk】Wire organizer under desk built-in soft rubber pads on the clamp protect your desk from scratches, so you will not damage your furniture or office.
- 【Sturdy and Beautiful】Cord management under desk is made of high-quality carbon steel and holds about 15 lbs. The wire basket has 2 holes on each side of the tray to enable your cables to pass freely. Cable baskets are suitable for desk thickness use from 0.4" to 2.4".
- 【Space Saving & Keep Organized】Desk cable management tray holds all your power cords, outlet strips, USB hubs and computer transformers hidden from your feet, and the data cables will not enter the vacuum cleaner to keep your desk clean and tidy. Desk wire organizer is also suitable for use in the kitchen and outdoors.
- 【Reduce Safety Risks】Cable tray under desk keeps all plugs away from your kids, no more worry about tripping. Using 2 holes wire tray saves extra space, helps you easily cabling and wire protection. The mesh design is not easy to accumulate dust, but avoids safety accidents caused by messy wires.
Why can a fast endpoint return old data?
Round-trip time and slot freshness are different properties. A node may respond quickly while its ledger view is behind the cluster. Compare returned slots with a healthy reference using the same commitment and track the lag over time.
Do HTTP results apply to WebSocket subscriptions?
No. Test WebSocket handshakes, notification delay, disconnects, reconnect duration, and recovery of missed notifications separately. HTTP p95 cannot establish subscription reliability.
Are public Solana endpoints suitable for production?
They are shared, rate-limited infrastructure intended for development. For production, evaluate private RPC access for capacity, support, rate limits, regional performance, freshness, and a tested fallback design.
Frequently Asked Questions
Is ping enough to measure Solana RPC latency?
No. Ping measures an ICMP path, while an RPC call includes HTTP connection behavior, provider routing, queueing, JSON-RPC execution, and response transfer. Measure the actual POST request and the method your application uses.
Which Solana RPC endpoint is fastest from my location?
There is no universal answer. Test candidates from the machine or region that will send production traffic, then compare median, p95, p99, failures, timeouts, and slot freshness.
How many requests should a latency test send?
One hundred requests is a practical starting point, not a universal requirement. Sample long enough to cover the operating period that matters and repeat under representative peak load without exceeding endpoint limits.
Why can a fast endpoint return old data?
Round-trip time and slot freshness are separate. Compare returned slots with a healthy reference at the same commitment and track any lag over time.
Do HTTP results apply to WebSocket subscriptions?
No. Measure WebSocket handshakes, notification delay, disconnects, reconnect time, and recovery of missed notifications separately.
Are public Solana endpoints suitable for production?
They are shared and rate-limited development infrastructure. Production applications should evaluate private RPC capacity, support, limits, regional performance, freshness, and fallback behavior.
The Bottom Line
Run the real JSON-RPC workload from the region where your application runs, repeat it, and compare median and tail latency with failures, timeouts, and slot freshness. Test transaction and WebSocket workflows separately, and choose production RPC capacity based on measured regional behavior rather than a single quick response.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Free tools Windows power users keep installed
One-click scans. No signup required.
