2016年7月29日

憑借其流體欺騙技術，新的安全供應商旨在檢測並幫助阻止網絡內的攻擊。在經曆了兩年的隱身模式後，安全供應商Acalvio於7月26日正式亮相，揭示了其欺騙技術如何檢測並幫助阻止攻擊者。領導Acalvio的是首席執行官Ram Varadarajan，他曾在1997年創立Arcot Systems，並於2010年以2億美元的價格將公司出售給CA，並在2015年之前擔任該公司的總經理。Varadarajan告訴eWEEK:“我之前的工作是通過加強認證和控製，讓人們遠離企業。”“這一次，我正在尋找惡意活動，一旦它已經越過企業邊界。作為一個比喻，Varadarajan說，在CA和Arcot，他關心的是鎖和門，而Acalvio更類似於房子內部的運動檢測。迄今為止，在兩年的隱形開發中，Varadarajan已經從Accel Ventures、Ignition Partners和Eileses capital等投資者那裏籌集了1700萬美元的風險投資。這筆資金將用於開發技術，並使該公司能夠進入市場。市場上已經有很多基於欺騙的安全技術，包括TrapX、Attivo和Illusive。瓦拉達拉詹指出，安全領域總體上是一個擁擠的領域，他說他想確保籌集到足夠的資金來開發一款與眾不同的產品。 Among the core innovations behind Acalvio is the idea of fluid deception. Varadarajan said that rather than running through logs looking for suspicious events, he wanted to invert the model and have the anomaly announce itself. Acalvio makes use of a combination of machine learning and data science to solve security problems.A key challenge with deception technology is having a scalable system that still provides high-quality results, Varadarajan said. To solve the challenge of quality at scale, Acalvio is introducing the concept of fluid deception. "The response to a malicious probe of a network is intelligently guided," he said.So, for example, if an attacker sends a network ping request, the network will still respond with the right information to make the ping request successful. But when the attacker then attempts to remotely log in via SSH (Secure Shell), the Acalvio system will start a full virtual machine in order to enable the SSH request. Fluid deception allows for dynamic responses, and it's based on advanced machine learning algorithms. "We're trying to hit the nail with appropriate hammer, rather than hitting all nails with the same hammer," Varadarajan said.Going a step beyond just enabling an organization to detect that an attacker is in the network, Acalvio also aims to help remediate security risks. Varadarajan said that his goal is to enable organizations to detect, engage and respond to threats. "Detection alone isn't enough. Bad guy will keep coming up because there is no penalty for trying again," he said. "Just shooing away attackers isn't enough. You have to learn from the interaction so that you can inoculate yourself from whatever the bad guy was trying to do. "Sean Michael Kerner is a senior editor at eWEEK and InternetNews.com. Follow him on Twitter @TechJournalist.