Back in August, BitMEX officially migrated its data infrastructure to AWS Tokyo from AWS Dublin. The move was intended to improve liquidity on the platform by reducing latency hurdles and thus increase the number of market-making participants on BitMEX.
The migration took place on 23 August at 02:00 UTC and was completed within just 10 minutes by our engineering team. Soon after, the platform was brought back in Cancel-Only mode at 03:00 UTC, cutting the whole process to under an hour.
Fast forward to today, exactly one-month post-migration, we are excited to share the noticeable impact of our move to AWS Tokyo. This piece is also a tribute to our engineering team – for completing a hefty move in the most unnoticeable way possible.
If you haven’t signed up for a BitMEX account yet, we’re currently offering $5,000 worth of rewards to new users – you can register here.
A big part of the migration success is thanks to the hard work of our engineering team. They’ve pulled long nights and weekends for months to ensure the process was executed in the most efficient way possible.
From the beginning, our team took a disciplined approach by building a parallel infrastructure from the ground up. Over 100 failover tests were conducted to resolve potential bugs and expose all the technical debt before making the final datacentre switch. This required massive collaborative effort from the tech teams to unify our systems and seamlessly migrate a significant volume of historical data in a record time.
What may seem like a minimal lift from the outside was, in fact, a precise and meticulously planned execution.
The core objective behind the data infrastructure migration was to align with Asia Pacific. This achieves several goals to serve our users:
Soon after we completed the migration in August, we shared some noticeable improvements on our liquidity. Since then, the liquidity on our major contracts (XBTUSDT and ETHUSDT) and others (XRPUSDT, SOLUSDT, HYPEUSDT etc) have improved measurably, as shown in the graphs below.
The graph below shows the liquidity within 5bps from April 2025 (pre-migration) to September 2025 (one-month post-migration) for XBTUSDT.
The graph below shows the liquidity within 5bps from April 2025 (pre-migration) to September 2025 (one-month post-migration) for ETHUSDT (green), XRPUSDT (grey), and SOLUSDT (purple).
The graph below shows the liquidity within 5bps from April 2025 (pre-migration) to September 2025 (one-month post-migration) for DOGEUSDT (red), PEPEUSDT (pink), SUIUSDT (blue), BCHUSDT (green), and HYPEUSDT (grey).
From April 2025 to date, we have achieved more than 185% increase in liquidity across nine perpetual contracts. This growth will continue as we finetune internal processes and onboard even more market-making participants.
The table below shows an overview of the exact liquidity improvements for each contract shown above.
|
Contract |
Improvement in liquidity within 5bps from April to September 2025 (%) |
|
185.00% |
|
|
401.85% |
|
|
8,632.99% |
|
|
10,116.20% |
|
|
2,913.99% |
|
|
14,132.60% |
|
|
2,829.53% |
|
|
3,349.65% |
|
|
10,256.60% |
In short: we’ve proven ourselves as the home of the OG Bitcoin perpetual swap XBTUSD. Now we can proudly say that BitMEX is on its way to becoming a serious venue for trading perpetual contracts for tokens beyond Bitcoin.
One of the biggest challenges of our datacentre migration was closing the physical latency gap. While we had the migration sorted within 1 hour of platform downtime, getting our intra-stack latency back to its pre-migration point was another hurdle.
We are now proud to share that our key latency metrics are stronger than before the migration. This was achievable through iterations with our key clients, technology partners, and of course – the hard work of our engineering team.
The graph below shows the end-to-end order placement latency, with the x-axis as time and the y-axis as latency. The green dotted line represents the datacentre migration date. The red line shows median latency while the blue line shows the 20th percentile latency.
Another significantly positive impact of the migration was the decrease in inter-venue latency to major markets in Tokyo, which helps our High Frequency Trading (HFT) clients and liquidity providers to increase our orderbook liquidity even more than before.
The graph below shows the maximum orderbook latency at the time of the datacentre migration, with the x-axis as time and y-axis as latency (m/s).
While we’re proud that the team has handled the migration seamlessly, we’re focused on continuing to upgrade and enhance our platform to handle what comes next. The crypto trading landscape is only growing – as it does, we will continue to enhance our reliability, capacity, and performance to provide our users with the best possible trading experience.
If you are a user that was affected by BitMEX’s liquidity or latency in the past, our message is simple – we have started and will continue to address this. We encourage you to see the results for yourself.
To be the first to know about our new listings, product launches, giveaways and more, we invite you to join one of our online communities and connect with other traders. For the absolute latest, you can also follow us on Twitter, or read our blog and site announcements.
In the meantime, if you have any questions please contact Support who are available 24/7.
The post こんにちは (“Konnichiwa”) Tokyo, Slán Dublin: BitMEX Has Migrated to AWS Tokyo appeared first on BitMEX Blog.
Also read: Ryanair passe au tout-dématérialisé : voici quand vos tickets papiers deviendront obsolètes
