Metasploit eyeing Linux and usability improvements; iOS support uncertain

Image: iStock/weerapatkiatdumrong


Metasploit owner Rapid7 is working on making its penetration testing software easier to use, more welcoming for Linux-based techniques, and a better partner to network security controls.

By Evan Koblentz | October 14, 2016, 10:50 PM PST


Engineers at Rapid7, which owns the popular Metasploit penetration testing tool, are preparing a variety of enhancements for the ramp-up to version 5.0 in 2017.

Metasploit evolved in 2003, Rapid7 acquired it from the original developers in 2009, and fourth-generation software debuted in 2011. Metasploit Pro is currently in version 4.2 and costs several thousand dollars for a license; Metasploit Framework currently in version 4.12.33 is open source, officials explained.


Leo Varela, director of engineering, said his team is developing capabilities such as a single-pane interface, ways to convert Android vulnerabilities into corporate network access, a new focus on automated testing of network security controls, and a code base that's slimmer and faster.


Metasploit is traditionally Windows-centric. However, for Apple iOS testing, Boston-based Rapid7 is in the same boat as everyone else in the security and forensics fields—it's very difficult to do. Varela said he's open to adding iOS modules if the community of open-source Metasploit Framework users can help. Apple's mobile operating system is a custom version based on a derivative of Unix, and in recent and upcoming changes, "We are adding the capabilities to be able to interact with Linux and with Unix," Varela noted.

"It's up to the open-source developers to add content to it. We believe these [other] investments are much more valuable to the penetration testing community at large while we allow the open-source community to come up with iOS modules," Varela added.

Joshua Marpet, of security and forensic consulting firm GuardedRisk, said Rapid7's ease-of-use plans sound helpful for lower-level employees, but security professionals are happy using the command line and would rather see Rapid7 put its resources into new modules.

Marpet gave an example of the recent distributed denial-of-service against prominent security blogger Brian Krebs. By going through network-connected street cameras, the attackers made whole new approaches, he said. That differs from the antivirus world where new viruses are typically just different payloads wrapped in existing techniques, he observed. Rapid7 needs to keep up with this, he urged.


Marpet, in Wilmington, Del., said another tool he likes is Strategic Cyber'sCobalt Strike because of its automation features. Washington D.C.-based developer Raphael Mudge made Cobalt Strike atop Metasploit Framework but later changed its foundation to a different system. Mudge, asked about his product's roadmap, said he has new releases every few months but declined to comment because of frequently changing priorities.

Unlike PlayStation 4 Pro, the Xbox Scorpio will natively support 4K: Microsoft exec

#CONSOLE WARS

Microsoft has confirmed that at least some of the games developed for the Xbox Scorpio will natively support 4K resolutions.

Referring to games developed by Microsoft for the Xbox Scorpio, Microsoft Studios publishing manager Shannon Loftis told USA Today, “Any games we’re making that we’re launching in the Scorpio time frame, we’re making sure they can natively render at 4K.”

The recently announced PlayStation Pro console was expected to support 4K as well, but Sony has since confirmed that the console will only upscale games to 4K rather than render them natively. Sony claims to be using advanced algorithms to achieve this.

The Xbox Scorpio is expected to arrive in “Holiday 2017,” giving the PlayStation 4 Pro a huge lead in the market. The PlayStation 4 Pro will be available for purchase starting 10 November this year at a price of $400 (around Rs 27,000).

However, Microsoft claims that the Xbox Scorpio will be almost 50 percent more powerful than the PlayStation 4 Pro (six TFLOPS vs. 4.14 TFLOPS). It’s clear from Loftis’ statement that the Xbox Scorpio will be capable of 4K gaming, but it’s also clear that other developers don’t necessarily have to support 4K.

Also, did you know that your PlayStation 4 now supports HDR?

Tags: game consoles, Microsoft, PlayStation, PlayStation 4, PlayStation 4 Pro, PS4, Scorpio vs Pro, Sony, video games,Xbox, Xbox One, Xbox Scorpio

Google self-driving cars may reduce fatalities, but what about human error

The week began with reports about Google’s autonomous car being involved in what it calls its worst ever accident. And, a human is to be blamed.

So, a van ran into a red light and rammed into the driverless car, damaging the right door and window. You can see the photo here. The Interstate Batteries van was to be blamed as the light is said to have been green for at least about ‘six seconds’ before the Google car entered the intersection.

This is neither the first time that a Google car is involved in an accident, nor is it the first company testing autonomous car to be facing the wrath. In the recent past we’ve seen Tesla involved in an accident that killed the driver. A DVD Player was found in the car that was on autopilot, and a truck rammed into it killing the driver. Again, a possible human error. However, some witnesses claimed that there was no movie or music playing in the car.

Earlier this year, Google car had hit a bus. The company did take up some responsibility for the hit. “We saw the bus, we tracked the bus, we thought the bus was going to slow down, we started to pull out, there was some momentum involved,” Chris Urmson, the head of Google’s self-driving car project, had told The Associated Press.

Can one rely on software guided cars?

If you read how the accident unfolded, one could never say who is at fault. It was plain ‘predicting the other’s move’ gone wrong. Google then said its computers had reviewed the incident and engineers changed the software that governs the cars to understand that buses may not be as inclined to yield as other vehicles.  However, in certain cases, who takes the responsibility of the accident? This also means users may not trust a software driving them around that won’t really take any responsibility.

In June, we heard how Google software is designed to see a 360 degrees view, improvement in honking algorithms so that it is more human like. Read here to know more. But, on a larger scale, the algorithm is still software that can fail or go wrong. And, the new worries on the forefront show how the car could rather foresee or avoid human errors. For example, in the recent case, the car entered the intersection after the light turned green, but it was the van that tried to zip in the last few seconds.

Predicting human error

So, unlike previous cases, wherein there was the chance of a system failure, bad prediction or likewise, this is a new worry for autonomous car makers. How can the smart car accurately foresee a human error? This also proves we are a long way to go until these cars take on drivers completely. Though driverless car is the most sought after category these days with major interest from global players including and not limited to Tesla, Google, Uberand nuTonomy, it may take a few years for a refined technology

In support of its autonomous cars, Google has been talking about how it will reduce fatalities caused by human error. Google said it  aims to develop a fully self-driving technology to make roads safer. But, here is the case of a smart car crash due to a human error. We wonder how long it would take to build a system that could cut through human errors and is designed to predict the move accurately

Nvidia's Tesla P4 And P40 GPUs Boost Deep Learning Inference Performance With INT8, TensorRT Support

Nvidia Tesl P40Nvidia continues to double down on deep learning GPUs with the release of two new “inference” GPUs, the Tesla P4 and the Tesla P40. The pair are the 16nm FinFET direct successors to Tesla M4 and M40, with much improved performance and support for 8-bit (INT8) operations.

Deep learning consists of two steps: training and inference. For training, it can take billions of TeraFLOPS to achieve an expected result over a matter of days (while using GPUs). For inference, which is the running of the trained models against new data, it can take billions of FLOPS, and it can be done in real-time.

The two steps in the deep learning process require different levels of performance, but also different features. This is why Nvidia is now releasing the Tesla P4 and P40, which are optimized specifically for running inference engines, such as Nvidia’s recently launched TensorRT inference engine.

Advertisement

Unlike the Pascal-based Tesla P100, which comes with support for the already quite low 16-bit (FP16) precision, the two new GPUs bring support for the even lower 8-bit INT8 precision. This is because the researchers have discovered that you don’t need especially high precision for deep learning training.

The expected results will appear significantly faster if you use twice as much data with half the precision. Because inference operates on already-trained data, even less precision is needed than for training, which is why Nvidia’s new cards now have support for INT8 operations.

Tesla P4

The Tesla P4 is the lower-end GPU from the two that were announced, and it’s targeted at scale-out servers that want highly-efficient GPUs. Each Tesla P4 GPU uses between 50W and 75W of power, for a peak performance of 5.5 (FP32) TeraFLOP/s and 21.8 INT8 TOP/s (Tera-Operations per second).

Nvidia compared its Tesla P4 GPU to an Intel Xeon E5 general purpose CPU and alleged that the P4 is up to 40x more efficient on the AlexNet image processing test. The company also claimed that the Tesla P4 is 8x more efficient than an Arria 10-115 FPGA (made by Altera, which Intel acquired).

Tesla P40

The Tesla P40 was designed for scale-up servers, where performance matters most. Thanks to improvements in the Pascal architecture as well as the jump from the 28nm planar process to a 16nm FinFET process, Nvidia claimed that the P40 is up to 4x faster than its predecessor, the Tesla M40.

The P40 GPU has a peak performance of 12 (FP32) TeraFLOP/s and 47 TOP/s, so it’s about twice as fast as its little brother, the Tesla P4. Tesla P40 has a maximum power consumption of 250W.

TensorRT

Nvidia also announced the TensorRT GPU inference engine that doubles the performance compared to previous cuDNN-based software tools for Nvidia GPUs. The new engine also has support for INT8 operations, so Nvidia’s new Tesla P4 and P40 will be able to work at maximum efficiency from day one.

In the graph below, Nvidia compared the performance of the Tesla P4 and P40 GPUs while using the TensorRT inference engine to a 14-core Intel E5-2690v4 running Intel’s optimized version of the Caffe neural networking framework. According to Nvidia’s results, the Tesla P40 seems to be up to 45x faster than Intel’s CPU here.

So far Nvidia has been comparing its GPUs to Intel’s general purpose CPUs alone, but Intel’s main product for deep learning is now the Xeon Phi line of chips with its “many-core” (Atom-based) accelerators.

Nvidia’s GPUs likely still beat those chips by a healthy margin due to the inherent advantage GPUs have even over many-core CPUs for such low-precision operations. However, at this point, comparing Xeon Phi with Nvidia’s GPUs would be a more realistic scenario in terms of what their customers are looking to buy for deep learning applications