EcrioniX/ Protocols/ CCIX Protocol
Interconnect Protocol

CCIX Protocol Explained
Cache Coherent Interconnect for Accelerators

How CCIX enables CPU–accelerator cache coherency over PCIe — architecture, transaction types, coherency model, and comparison with CXL.

PCIe
Physical Layer
MOESI
Coherency States
ARM CHI
Protocol Basis
HOST CPU CPU Cores + L1/L2 Cache L3 Cache (Shared) CCIX Home Node (HN) Memory Controller CCIX LINK (PCIe Gen3/Gen4 Physical Layer) ACCELERATOR (GPU / FPGA / ML ASIC) CCIX Request Node (RN) Local Cache (RN-F) Compute Engine Device Memory (optional) CPU and Accelerator share coherent view of memory — no software cache flush required

What is CCIX? OVERVIEW

CCIX (Cache Coherent Interconnect for Accelerators) is an open industry standard protocol that adds hardware cache coherency between host CPUs and attached accelerators over a PCIe physical link. Developed by the CCIX Consortium (founded 2016 by AMD, ARM, Huawei, IBM, Mellanox, Qualcomm, Xilinx), it solves the fundamental problem of heterogeneous computing: keeping data consistent between CPU caches and accelerator caches without software intervention.

Without coherency, a programmer must explicitly flush CPU caches before the accelerator reads the data, and flush accelerator memory before the CPU reads results. CCIX eliminates this overhead in hardware, enabling zero-copy data sharing.

Key insight: CCIX = PCIe physical layer + ARM CHI-like coherency protocol on top. It reuses the existing PCIe infrastructure (same connectors, same PHY) and adds a new Transaction Layer that carries coherency messages — no new hardware slots needed.

CCIX vs CXL vs CAPI COMPARISON

PropertyCCIXCXLCAPI (IBM)
Founded2016, CCIX Consortium2019, Intel-led2013, IBM
Physical layerPCIe Gen3/Gen4PCIe Gen5/Gen6PCIe Gen3+
Protocol basisARM CHI-likeCXL.cache, CXL.mem, CXL.ioIBM POWER architecture
Coherency modelFull MOESI snoop-basedFull coherency (CXL.cache)Full coherency
Memory semanticHost memory + device memoryCXL.mem for device memoryHost memory
Industry status (2026)Legacy / limited new adoptionDominant standardIBM Power-only
Key adoptersAMD EPYC, Xilinx FPGAs, ARM serversIntel, AMD, NVIDIA, Samsung, all major SoCIBM Power servers

Industry shift: CXL has become the dominant cache-coherent interconnect standard. The CXL Consortium (which absorbed many CCIX members) has broader industry backing and integration into PCI-SIG. New designs in 2024+ are almost exclusively using CXL. CCIX remains relevant for understanding existing AMD EPYC and ARM server deployments.

CCIX Architecture — Key Nodes ARCHITECTURE

Node TypeRoleExample
RN-F (Request Node - Full)Fully coherent agent — has a cache, participates in snoopsGPU with cache, FPGA compute engine
RN-I (Request Node - I/O)Non-caching agent — issues read/write without snoopsDMA engine, I/O device
HN-F (Home Node - Full)Point of coherency — receives requests, issues snoops, orders transactionsCPU LLC controller, SoC NIC
HN-I (Home Node - I/O)Manages I/O address space, non-coherentPeripheral fabric controller
SN (Slave Node)DRAM controller — serves data to HNDDR5 memory controller

CCIX Transaction Types PROTOCOL

TransactionDirectionCoherencyPurpose
ReadNoSnoopRN → HNNon-coherentDMA-style read, no cache involvement
ReadOnceRN → HNCoherent, transientRead data once, don't cache long-term
ReadSharedRN → HNCoherent SharedRead and cache in Shared state
ReadUniqueRN → HNExclusive (write intent)Read with intent to modify — invalidates other copies
MakeUniqueRN → HNExclusive (upgrade)Upgrade Shared → Exclusive without data transfer
WriteNoSnoopRN → HNNon-coherentNon-coherent write to memory
WriteUniqueRN → HNCoherent writeWrite to unique copy — invalidates Shared copies
EvictRN → HNCache managementNotify HN that a clean Shared line is being evicted
SnpSharedHN → RNSnoopHN-initiated: downgrade cache line to Shared
SnpUniqueHN → RNSnoopHN-initiated: invalidate cache line (for new exclusive owner)
SnpCleanInvalidHN → RNSnoopHN-initiated: writeback dirty data and invalidate

Cache Coherency States (MOESI) COHERENCY

StateMeaningCan Read?Can Write?Must Writeback?
M — ModifiedOnly copy, dirty (differs from memory)YesYesYes (on eviction)
O — OwnedDirty, shared with others — owner must supply data on snoopYesNoYes
E — ExclusiveOnly copy, clean (matches memory)YesYes (silent → M)No
S — SharedClean, multiple caches may holdYesNo (must upgrade)No
I — InvalidNot present in cacheNo (must fetch)NoNo

CCIX Use Cases APPLICATIONS

CCIX Protocol Stack LAYERS

LayerCCIX DefinitionStandard Equivalent
ApplicationCoherent memory transactionsCustom per use case
Transaction LayerCCIX TLP extensions (over PCIe TLP)PCIe TLP + CCIX header
Data Link LayerPCIe DLLP (unchanged)PCIe standard
Physical LayerPCIe Gen3/Gen4 SerDesPCIe standard

How CCIX reuses PCIe: CCIX negotiates capability during PCIe link training using an extended capability structure. When both endpoints support CCIX, the link switches a portion of its bandwidth to carry CCIX TLPs alongside standard PCIe TLPs. No new connectors or cables are required.

Frequently Asked Questions FAQ

What is CCIX protocol? +

CCIX (Cache Coherent Interconnect for Accelerators) is an open standard that extends PCIe with hardware cache coherency between CPUs and accelerators (GPUs, FPGAs, AI chips), eliminating software-managed cache flushes for shared data.

What is the difference between CCIX and CXL? +

Both add coherency over PCIe. CCIX (2016) was ARM/AMD-led; CXL (2019) is Intel-led and now the dominant standard with full PCI-SIG integration. CXL has three sub-protocols (CXL.io, CXL.cache, CXL.mem) and broader industry adoption. New designs in 2024+ use CXL.

What chips support CCIX? +

AMD EPYC Rome/Milan CPUs, Xilinx Alveo FPGAs (U280, U250), Marvell ThunderX2, Ampere Altra ARM servers, and various AI accelerator ASICs. Many are shifting to CXL for next-generation designs.

How does CCIX coherency work? +

CCIX uses a snoop-based MOESI protocol. The CPU's Home Node (HN) tracks which nodes cache which lines. When an accelerator requests a cache line, the HN snoops CPU caches, forces writebacks if needed, then supplies data — all in hardware, transparent to software.

Is CCIX still relevant in 2026? +

CCIX remains relevant for understanding existing AMD EPYC server deployments and Xilinx FPGA accelerator platforms. For new designs, CXL has largely replaced it. Understanding CCIX is still valuable for SoC architects and VLSI engineers working with heterogeneous compute.

Related protocols:

All Protocols DDR Protocol VLSI What is a GPU