Connect with us
New IBM processor breaks 100 qubits of useful computing New IBM processor breaks 100 qubits of useful computing

Security

New IBM processor breaks 100 qubits of useful computing

Published

on

The company intends to release a processor covering more than 1,000 cubits in 2023.

IBM company overcame a major obstacle to the usefulness of quantum computing with the development of the Eagle quantum processor with 127 functional qubits. Thus, IBM became the first company to break the 100-qubit mark, which is an interesting milestone because the interaction of so many qubits cannot be modeled using modern classical computing hardware and algorithms.

According to Darío Gil, director of research at IBM, the company intends to develop the first 1,000-qubit processor over the next two years. According to Gil, the new qubit record is the result of several developments that have been combined for the first time. First, IBM switched to a so-called “heavy hexadecimal” qubit arrangement, which combines qubits into a set of hexagons with common sides. This approach allows qubits to be less likely to influence their neighbors, leading to errors.

While the qubits remain at the same level within the processor, the processor as a whole uses additional layers to place signal wires that allow the qubits to be driven and read. This is common in traditional microcircuits, but new to the world of quantum computing. The wires can also carry multiplexed signals.

Other important advances included the ability to tune the frequencies of the microwaves to which each individual qubit responds, preventing so-called “collisions” in which a signal destined for one qubit could also change the behavior of others. According to Gil, a quantum “compiler” sets up a sequence of operations that perform computations on quantum hardware and can also help avoid “collisions.”

Finally, IBM reduced the error rate of the main function block of the two-qubit gate to 0.001. This is very important, because if the error rate remains constant as the number of qubits increases, a large amount of computation is likely to be accompanied by errors.

Click to comment

Leave a Reply

Your email address will not be published.

Components

Vulnerability found in the Apple M1 processor that cannot be closed programmatically

Published

on

Vulnerability found in the Apple M1 processor that cannot be

Specialists from the Computer Science and Artificial Intelligence Laboratory (CSAIL) from the Massachusetts Institute of Technology said that they were able to develop a PACMAN attack technique, which is possible due to a hardware vulnerability in the Apple M1 processor. Moreover, it cannot be closed by software.

Vulnerability found in the Apple M1 processor that cannot be closed programmatically

The attack itself is carried out using software and hardware, and it can be performed remotely. The PACMAN attack makes it possible to gain access to the kernel of the operating system. At the same time, potentially the same vulnerabilities can be in Qualcomm and Samsung processors, but this has not yet been confirmed.

The technical side of the attack is based on the Pointer Authentication function. It is used to check Pointer Authentication Codes (PACs), allowing only code-signed software to run. However, the PACMAN method allows you to select the necessary PAC values. In part, this technique is similar to Specter and Meltdown in Intel processors.

At the same time, Apple spokesman Scott Radcliffe said that the vulnerability does not pose an immediate threat to users and is not sufficient in itself to bypass operating system protections.

A few years ago, a vulnerability was already found in Arm processors that allows an attacker to gain unauthorized access to data.

Continue Reading

Security

So much for Unisoc. Companies have discovered a vulnerability in single-chip systems

Published

on

So much for Unisoc Companies have discovered a vulnerability in

Unisoc is actively capturing the market for single-chip systems, although it does so exclusively in the budget segment itself. However, it turned out that these platforms have a critical vulnerability.

So much for Unisoc.  Companies have discovered a vulnerability in single-chip systems

According to the source, the problem is in the modem’s firmware and affects both 4G and 5G platforms. The vulnerability, numbered CVE-2022-20210, was discovered while scanning Non-Access Stratum (NAS) message handlers. This vulnerability could be used to neutralize or block the device’s cellular capabilities.

The vulnerability was first discovered in the Motorola Moto G20 smartphone based on the Unisoc T700 SoC. But in the end it turned out that the same vulnerability occurs in other platforms, however, the source did not provide a list.

The Check Point specialists who discovered the vulnerability notified Unisoc back in May, and the company has already released a fix, so smartphone owners should not worry now if they update the software of their devices.

Continue Reading

Security

New vulnerability in Microsoft Office is heavily exploited by hackers

Published

on

New vulnerability in Microsoft Office is heavily exploited by hackers

A serious vulnerability has been found in the Microsoft Office suite that could potentially allow attackers to execute arbitrary code.

New vulnerability in Microsoft Office is heavily exploited by hackers

She was assigned the number CVE-2022-30190, and among the researchers they gave the name Follina. As noted in Kaspersky Lab, the most unpleasant thing is that there is no fix yet, and in the meantime, the vulnerability is already being actively exploited by attackers. Vulnerability CVE-2022-30190 threatens all operating systems of the Windows family, both regular and server.

While the update is being developed, experts recommend that all Windows users and administrators take advantage of temporary workarounds. As a Microsoft workaround recommends disable the MSDT URL protocol.

The CVE-2022-30190 vulnerability itself is contained in the Microsoft Windows Support Diagnostic Tool (MSDT), but due to the implementation of this tool, a single malicious office document is enough to exploit the vulnerability.

The scheme could be like this. Attackers create a malicious office document and slip it to the victim, for example by sending an email with an attachment. The infected file contains a link to an HTML file that contains JavaScript code that executes malicious code on the command line via MSDT. As a result, attackers are able to install programs, view, modify or destroy data, as well as create new accounts – that is, do everything that the privileges of the user who opened the infected file allow.

Continue Reading

Most Popular