Clockwork Latency Sensei helps servers prepare for real-time applications

Hello and welcome to Protocol Enterprise! Today: how a new startup plans to solve an age-old business tech problem, cloud offices might finally be an idea that sticks, and three data scientists walk into a bar: this AI can predict what happens next.

Twirl up

Site Reliability Engineers (SREs) are the people who keep the cloud running and tend to only get noticed for their efforts when something goes wrong. Perhaps that’s why 85% of SREs are turning to automation and AIOps services for help as their business evolves. according to Dynatrace.

No one told you when to run

The people who know computers best consider it a miracle that they actually work. A new business startup called Clockwork wants to solve an old fundamental problem with computer networks – their inability to keep very accurate time.

Its mission is to bring nanosecond clock synchronization accuracy in distributed systems to empower time-sensitive applications used in cryptocurrency and stock trading, mobile banking, online gaming, database design, and other industries.

  • “If you look at cloud computing, the big idea of ​​the first decade of the century was virtualization,” said Clockwork co-founder and CEO Balaji Prabhakar. “The next decade was the decade of big data, dealing with large amounts of data. What we think is that this decade is going to evolve in time, speed, deadlines, real-time control: things to be done with more time sensitivity.
  • Latency Sensei is Clockwork’s new network latency sensor for cloud, hybrid and on-premises data center environments.
  • Leveraging the company’s clock synchronization technology, the software measures one-way network delays and helps customers detect bottlenecks, misfires, and underperforming virtual machines to optimize application performance. .

Company Executives Say Sensei’s Latency Sensor Breaks Through “Virtualization Fog” to give DevOps engineers visibility into the underlying infrastructure of their networks.

  • “When you go into the cloud and rent virtual machines, you’re in a bubble,” Prabhakar said.
  • “You don’t know what hardware you’re working on, you don’t know if two virtual machines are in the same server rack or are they next to each other in the data center. And when you don’t know, you can’t tell how long you’ve been on the network.
  • Latency Sensei promises to solve this problem, determining as accurately as possible how long it takes for a packet or any piece of data to travel from one node in a network to another node.
  • “For 50 years…in networking, we’ve never really been able to accurately measure one-way delays because we don’t have accurate clocks,” Prabhakar said, referring to unsynchronized network clocks. . “All we’ve done historically is say, ‘The time from here to there is half the time it takes to get from here to there and back. “”
  • This round-trip time approach is not adequate for determining where a delay is occurring, as it does not distinguish between forward and backward congestion which can be detected with precise times at one way, he said.

Clockwork launched in 2018 to commercialize clock synchronization research conducted at Stanford University under the supervision of Prabhakar and VMware co-founder Mendel Rosenblum, who is Clockwork’s chief scientist.

  • Latency Sensei’s underlying time synchronization technology comes from Clockwork’s first software product, Clock Sync, which synchronizes computer clocks to extremely high levels of precision: precision to single digit nanoseconds for hardware timestamps and to hundreds nanoseconds for software timestamps, according to the company. .
  • Clock Sync software was released for private data centers in March 2019 and for public clouds a year later. It can scale up to thousands or tens of thousands of nodes.
  • Accurate synchronization of clocks is an old problem, and the unstable nature of the networks connecting the clocks in the servers makes it difficult to solve, as the networks could add random delays to the packets exchanged by the clocks, according to Prabhakar.
  • “The network is kind of the enemy in this equation,” he said. “The approach taken was to synchronize the network switches with the reference clock and use the network to transmit the time. This is a hardware-based method that is expensive and difficult to scale. »

Clockwork proposed a software approach which does not require a network upgrade.

  • “The way we did it was not to touch the network, but to make it more of a signal processing, machine learning type approach, just cleaning up the timestamps we get from all the random delay noise added by the network,” Prabhakar explained.
  • “We treat the network as a black box and make clock synchronization an application service. This makes us scalable and accurate. Could it have been done 30 or 25 years ago, probably not – not the way we did, because the ML technology we needed was not there.

“Having computers in sync is a really powerful thing,” said Greg Papadopoulos, NEA partner and former chief technology officer at Sun Microsystems.

  • “There is so much promise here. It’s really something you don’t see that often… where you can take such a fundamental idea that can transform a territory and, at the same time, you’re not really sure what it means. It will really be the story of how we taught computers to keep time.

— Donna Goodison (E-mail | Twitter)


DuckDuckGo offers an all-in-one privacy solution aimed at simplifying online privacy protection. DuckDuckGo’s app can be used as an everyday browser with private search, tracker blocking, encryption and now email protection built-in. This is the free and easy button for online privacy.

Learn more

Armonk on the mainline

Despite all the criticism leveled at mainframe computers at the start of the pandemic, which exposed how many US states rely on creaky and aging technological infrastructure for day-to-day operations, these machines continue to have remarkable battery life.

IBM unveiled the latest generation of its venerable Tuesday mainframe product line, called the z16. It runs on a custom IBM-designed processor, and while most mainframe customers use them for transaction processing, the new machine can also handle AI workloads.

“Despite the continued move to the cloud, IBM says two-thirds of the Fortune 100, 45 of the world’s 50 largest banks, eight of the 10 largest insurers, seven of the 10 largest global retailers and eight of the 10 largest telecom operators rely on its mainframes for critical processes,” according to Data Center Dynamicshighlighting the role that mainframes continue to play in enterprise technology.

But IBM couldn’t help but overplay its hand, saying the new systems will be “the industry’s first secure quantum system.”[s]implying that they will be able to withstand attacks from future quantum computers that have yet to be built. In the fine print under its press release, IBM attributed this claim to “a third-party analyst” who referred to List of NIST post-quantum cryptography algorithmswhich is really a work in progress.

And you will also need to upgrade to the Crypto Express 8S card to be quantum safe.

—Tom Krazit (E-mail | Twitter)

Coming to Protocol

It has been almost six months since Congress passed the landmark $1 trillion Infrastructure Investment and Jobs Act. What progress towards these goals have we seen so far – and what can we expect over the next six months?

In this Protocol virtual event on April 21 at 9 a.m. PT, we’ll explore how the infrastructure bill rollout is going and what it means for you. Join Protocol’s Issie Lapowsky in a conversation with Alan Davidson, Assistant Secretary for Communications and Public Information, US Department of Commerce; Nicol Turner Lee, Principal Investigator and Director of the Center for Technology Innovation, The Brookings Institution; and Angela Siefer, Executive Director, National Digital Inclusion Alliance. RSVP here.

What’s wrong with machine learning?

AI-based language models are far from perfect, but researchers continue to plug into machines that can replicate human communication skills. Humor is a very useful and entertaining communication skill, and Google researchers said this week they are progressing on a version of his Pathways Language Model (PaLM) that can explain jokes.

“Remarkably, PaLM can even generate explicit explanations for scenarios that require a complex combination of multi-step logical inference, knowledge of the world, and deep understanding of language. For example, it can provide high-quality explanations for new jokes not found on the web,” the researchers wrote. There’s a lot of work to do, but at least it’s an improvement over previous language models, who were themselves a joke.

—Tom Krazit (E-mail | Twitter)


Tracking is a global problem – over 80% of websites, apps and emails contain third-party trackers. For this reason, people need a multi-pronged privacy solution. DuckDuckGo’s all-in-one privacy app can be used as an everyday browser with several built-in features including private search, tracker blocking, encryption, and email protection.

Learn more

Thanks for reading – see you tomorrow!

Previous St. Mary's University Again Hosts Fiesta Jazz Festival
Next Take center stage with Bo Tracy | News, Sports, Jobs