Today’s classical supercomputers can do a lot. However, because their calculations are confined to binary states of 0 or 1, they struggle with really difficult tasks like natural science simulations. Quantum computers, which may represent information as 0, 1, or both at the same time, may be advantageous in this situation.
Last year, IBM unveiled a 127-qubit computer device as well as the IBM Quantum System Two, a structure that would store components such as the chandelier cryostat, wiring, and circuitry for future processors. In the battle to construct the most powerful quantum computer, these advancements placed IBM ahead of other prominent tech giants like Google and Microsoft. With a processor dubbed "Kookaburra," the business is putting out a three-year strategy to surpass 4,000 qubits by 2025. This is how it intends to get there.
IBM will continue to develop both the hardware and software components for quantum processors in order to expand its processing capabilities for qubits. The first is Heron, a revolutionary processor with 133 qubits. The Heron chip, in addition to having more qubits, is designed differently from its predecessor, Eagle.“It actually allows us to get a much larger fraction of functioning 2-qubit gates. It’s using a new architecture called tunable couplers,”explains Jerry Chow, IBM Quantum's head of quantum hardware system development.
“Along with this plan for this new processor for Heron, we want to be able to have multiple Herons that are all addressable via one control architecture,” he continues.“We want to be able to have classical communication linked across these chips and processors as we’re building them out.”
Better gate-level control
Before you can understand what a qubit is you need to understand what a bit and a gate are.
Information is encoded as binary bits on traditional computers (0 or 1). Transistors are electronic switches that regulate electron flow. A gate electrode is attached to a transistor. Changing the electrical charge on the gate electrode switches the transistor between states 1 and 0. Computers can encode information through physical changes to these states.
Logic gates are transistors arranged in a specified pattern. An integrated circuit is made up of transistors that can store data in portions. On the surface of a chip, these circuits are all coupled.
Quantum gates function differently than classical gates, and qubits work differently than bits. Unlike conventional bits, which can have a value of 1 or 0, qubits can persist in a wave-like quantum superposition state, which represents a combination of all conceivable configurations—0, 1, or both at the same time—under the correct conditions. Researchers can regulate the behavior of microwave photons by firing them at qubit-specific frequencies, which may be used to keep, modify, or read out units of quantum information.
Unfortunately, qubits are extremely delicate: they are heat-sensitive, unstable, and prone to errors. When qubits communicate with one another or with the wire in their environment, their quantum qualities can be lost, resulting in less precise calculations. Experts talk about their "coherence time" when explaining how long they can stay in superposition states. The size of a quantum calculation you can execute with a set of qubits is limited by the coherence time and the time it takes to perform a gate.
“The way that we’ve been designing our current processors, Falcon, Hummingbird, Eagle, have been using fixed coupling between qubits, and we’ve been using a microwave-based 2-qubit cross-resonance gate,”explains Chow.
They were communicating with the appropriate qubit on different frequencies in such circumstances. Now, according to Chow, they're adding "individualized magnetic field controls for the couplers connecting the qubits," which will allow them to turn on qubit interactions at different microwave frequencies.
Multiple, connected quantum processors
Cores are groups of transistors in traditional computers that may execute numerous operations in concurrently. Imagine having many checkout registers open at a supermarket rather of having everyone queue up for one. Multi-core CPUs, also known as multi-threading CPUs, may break down a large workload into smaller chunks that can be given to separate cores for processing.
IBM now intends to use circuit knitting to extend this approach to quantum computing. This “effectively takes large quantum circuits, finds ways to break them down into smaller, more digestible quantum circuits, which can be almost parallelly run across a number of processors” Chow notes. “With this classical parallelization, it increases the types of problems and capabilities that we’re able to address.”
Parallelization might also help reduce mistake rates.
This design offshoot is separate from the development of Osprey or Condor, which are on track to reach 433 and 1,121 qubits, respectively, in the next few years. “But we also want to have some modularity built-in that will allow us to scale even further. At some level, just the amount of the number of qubits that we’re going to be able to pack into a single chip will start to become limited,” adds Chow. “We’re testing some of those boundaries with Osprey and with Condor currently.”
Engineers will use Heron to evaluate methods for establishing quantum connections across numerous quantum chips.“We’re exploring what we call these modularly couplers that will allow us to effectively have multiple chips that are connected together,” Chow explains.
This will result in a bigger quantum coherent processor made up of three independent quantum chips sharing the same quantum processor. In order to do this, IBM plans to combine three processors into the Crossbill 408-qubit device in 2024.
IBM is also developing long-range couplers that can connect clusters of quantum processors through a meter-long cryogenic cable, allowing the company to grow much farther (superconducting qubits need to be kept very cold). “We’re calling this the inter-quantum communication link,” Chow adds, explaining that it can expand quantum coherent communications inside the common cryogenic environment.
IBM is also developing long-range couplers that can connect clusters of quantum processors through a meter-long cryogenic cable, allowing it to expand much farther (superconducting qubits need to be kept very cold). According to Chow, the inter-quantum communication link can expand quantum coherent connections inside the common cryogenic environment.
The Kookaburra, a 4,158-qubit system that combines parallelization, chip-to-chip communication, and long-range coupling, might help them meet their 2025 objective.
Combining classical computing with quantum computing
Going quantum does not necessitate a complete computer rebuild. The quantum system uses a lot of traditional computer infrastructure. “The way that we typically have our systems is you have your quantum processor inside the refrigerator and you’re constantly talking to it with the classical infrastructure,” Chow says.“The classical infrastructure is generating these microwave pulses, generating the read-outs. When you program a circuit it just turns into this orchestration of gates, operations that go to the chips.”
However, rather of having just quantum processors, one controller can also feed into traditional processors such as CPUs and GPUs, which would be linked in parallel to the quantum chip but not in any quantum sense. It will be able to run threaded programs using both conventional and quantum computing capabilities.
“The quantum processor is providing a different resource from a GPU or a super large CPU,” explains Chow. “But overall, the whole thing is going to be something that feels like a supercomputer that is still orchestrated together.”
Machine components will be able to execute quantum circuits on quantum hardware, according to IBM's vision of the future of computation. This component, on the other hand, will be sewn together with traditional memory and infrastructure. The variational quantum eigensolver is a hybrid quantum-classical technique that may be used to address issues like molecular simulations.
Quantum software
Classical circuits are not the same as quantum circuits. The gates have distinct logic, and the algorithms have different language.
When IBM's first quantum computer was released to the cloud in 2016, it came with an assembly language called OpenQASM that was used to create applications. IBM's OpenQASM 3 library will include "dynamic circuits" that can simultaneously measure qubits and process classical information this year. This is a hardware upgrade that relies on better control electronics and real-time communications between the control and measurement sides of the circuit. More mistake fixes and parity checks may be possible.
Primitives, or the basic compute pieces of an algorithm, will be part of IBM's Qiskit Runtime platform, a quantum computing service and programming paradigm. Qiskit has several levels of assembly languages for kernel developers who may need to interact with code and hardware, as well as a serverless API for algorithm developers.
“At this higher level for algorithm developers, you don’t need to care about running it on any particular backend when you have this cloud environment where you can access the CPUs, GPUs, and QPUs, all orchestrated together,” Chow adds.
“It allows us to use the classical resources in concert with our quantum resources to handle some of the larger quantum circuit problems—ones that might be pushing on things like quantum advantage.”
Tags:
technology