IBM just enjoyed its conference Quantum Summit 2021 to present its new breakthrough in quantum computing. After his processor Hummingbird of 65 qubits announced last year, the manufacturer unveiled Eagle, a quantum processor of 127 qubits.
Many companies are currently engaged in a real quantum computer race. Microsoft and Amazon already offer quantum computing services with Azure Quantum and Braket, respectively. However, these services are limited at the hardware level. Quantum processors have so far been blocked by the 100 barrier qubits, which IBM has just broken.
The manufacturer has just announced Eagle, the first 127 qubit quantum processor. As a reminder, the qubit is the quantum equivalent of the binary bit of classical computing. Where the bit can have a value of 0 or 1, the qubit can also be both simultaneously. IBM succeeded in creating this new processor thanks to two advances. The first is multiplexing, which eliminates the need to integrate control and read circuits into each qubit. The second is the integration in three dimensions, with the connections on different levels. physical, and a level dedicated only to qubits, more dense.
IBM presents its vision of a quantum data center with its future Quantum System Two, still at the concept stage. (Activate automatic translation of subtitles). © IBM
A quantum processor of 1,121 qubits within two years?
IBM also unveiled its plans for the future, with a 433 qubit Osprey processor scheduled for 2022, followed by a 1,121 qubit Condor processor in 2023. The manufacturer also announced Quantum System Two, a new design byquantum computer to replace the current one Quantum System One. It is still at the concept stage, but IBM wanted to consider a modular computer with denser cryogenic systems. The very large cryogenic elements are necessary for the quantum properties of the qubits to manifest. The manufacturer should present the first prototype of the Quantum System Two in 2023.
Selon IBM, Eagle is the first quantum processor to be too complex to be simulated on a conventional computer. Eagle should allow a new level of complexity in experiments, in order to optimize machine learning or model new molecules.
—
Discover TechPod, the bimonthly summary of tech and mobility news!
—
Interested in what you just read?
.