between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. Solo (deprecated in v2.0) The Solo implementation long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT

between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. 1.7 Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. • Solo (deprecated in v2.0) The Solo long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT

有效率的算法，采用硬件加速等。平行扩展指系统架构具有良好的可扩展 性，可以采用分片、分区的方式承载不同的用户、业务流的处理，只要适当 增加软硬件资源，就能承载更多的请求。 性能指标和软件架构，硬件配置如CPU、内存、存储规格、网络带宽都密切 相关，且随着TPS的增加，对存储容量的压力也会相应增加，需要综合考 虑。 安全性 安全性是个很大的话题，尤其是构建在分布式网络上多方参与的区块链系 统。在系统层 直接查看具有较高安全隐私等级的用户数据，交易数据。 系统运行过程中，可通过监控系统对各种运行指标进行监控，对系统的健康 程度进行评估，当出现故障时发出告警通知，便于运维快速反应，进行处 理。 监控的维度包括基础环境监控,如CPU占比、系统内存占比和增长、磁盘IO情 况、网络连接数和流量等。 区块链系统监控包括如区块高度、交易量和虚拟机计算量，共识节点出块投 票情况等。 接口监控包括如接口调用计数、接口调用耗时情况、接口调用成功率等。 生成的源码文档位于build/doc $ make doc 硬件要求 注解 由于节点多群组共享网络带宽、CPU和内存资源，因此为了保证服务的稳定 性，一台机器上不推荐配置过多节点。 下表是单群组单节点推荐的配置，节点耗费资源与群组个数呈线性关系，您 可根据实际的业务需求和机器资源，合理地配置节点数目。 配置 最低配置 推荐配置 CPU 1.5GHz 2.4GHz 内存 1GB 8GB 核心 1核 4核 带宽 1Mb

between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. Solo (deprecated in v2.0) The Solo implementation long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT

between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. 1.7 Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. • Solo (deprecated in v2.0) The Solo long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT

支持海量数据存储 并行计算 支持块内交易并行执行 节点类型 共识节点、观察节点 计算模型 排序-执行-验证 系统性能 峰值TPS 2万+ TPS（PBFT） 交易确认时延 秒级 硬件推荐配置 CPU 2.4GHz * 8核 内存 8GB 存储 4TB 整体架构 网络带宽 10Mb 账本模型 数据结构 链式结构，区块通过哈希链相连 是否分叉 不分叉 记账类型 账户模型（非UTXO） 有效率的算法，采用硬件加速等。平行扩展指系统架构具有良好的可扩展 性，可以采用分片、分区的方式承载不同的用户、业务流的处理，只要适当 增加软硬件资源，就能承载更多的请求。 性能指标和软件架构，硬件配置如CPU、内存、存储规格、网络带宽都密切 相关，且随着TPS的增加，对存储容量的压力也会相应增加，需要综合考 虑。 安全性 安全性是个很大的话题，尤其是构建在分布式网络上多方参与的区块链系 统。在系统层 直接查看具有较高安全隐私等级的用户数据，交易数据。 系统运行过程中，可通过监控系统对各种运行指标进行监控，对系统的健康 程度进行评估，当出现故障时发出告警通知，便于运维快速反应，进行处 理。 监控的维度包括基础环境监控,如CPU占比、系统内存占比和增长、磁盘IO情 况、网络连接数和流量等。 区块链系统监控包括如区块高度、交易量和虚拟机计算量，共识节点出块投 票情况等。 接口监控包括如接口调用计数、接口调用耗时情况、接口调用成功率等。

between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. 1.7 Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. • Solo (deprecated in v2.0) The Solo long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT

between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. Solo (deprecated in v2.0) The Solo implementation long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT

between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. Solo (deprecated in v2.x) The Solo implementation long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT

between members on a channel, allowing much of the same protection as channels without the maintenance overhead of creating and maintaining a separate channel. 1.7 Pluggable Consensus The ordering of transactions service has been available since Fabric v1.0, but many users may find the additional administrative overhead of managing a Kafka cluster intimidating or undesirable. • Solo (deprecated in v2.x) The Solo long as only a single node is added or removed at a time). While this configuration creates more overhead in the form of redundant heartbeat messages and goroutines, it lays necessary groundwork for BFT