Layered architecture Modular monoliths Microkernel SOA Service-based architectures Microservices ���� ���� FP ���� ���� MicroKernel MVC Layered Architecture Hexagonal Architecture Onion Architecture Architecture Screaming Architecture ����� �� �� ���� BDD ����������������������������George Fairbanks�Pattern-oriented Software Architecture����� �������� ��� + ���� ���� ���� ���� ���������� �����Problem Space� ����Solution Space� �����Architectural Pattern� �����Design Pattern� ���Architecture� ���� �� �Module� ����� �Bounded Context� �� �Component� �� �service� � �library�

API Cluster A Worker Worker Worker Kubernetes Master API Cluster B Operator Developer Architecture PKS Controller IaaS Worker Worker Kubernetes Cluster Services API Cluster3 NSX-T vSphere create-cluster K8s-1 n=3 #pks create-cluster K8s-2 n=3 #pks create-cluster K8s-3 n=3 #pks resize K8s-3 n=5 Architecture NSX-T Bosh PKS Admin Network NCP POD 1 POD 4 POD 2 POD 3 POD 5 POD 6 T0 kube-system Pod 6 T1 T1 T1 T1 VMware vSphere NSX Manager NSX Controllers T1 NSX Edge Cluster Architecture NSX-T • NSX Container Plugin: NCP is a software component provided by VMware in form of a

TensorFlow 和 PyTorch 和 Hugging Face Transformers 等。 TensorFlow 架构图 (图源：https://www.geeksforgeeks.org/architecture-of- tensorflow/) 12 / 32 LLM 基础设施：编程语言 LLM 的训练和应用通常使用多种编程语言，取决于任务的需求和团 队的偏好。 。它的广泛使用得 益于其简洁的语法、强大的库支持（如 LLM 基础设施：编程语言  2023 年 9 月面向大众开放 创业公司 Modular AI 开 发  结合了 Python 的易用性以及 C 语言的  支持与任意 代码  性能是 Python 的 倍 Mojo�与其他语言性能对比 （图源：https://www.modular.com/max/mojo） 15 / 32 大模型应用现状 2022 年底大模型应用

openEuler 21.09 Technical White Paper 3 Scenario-Specific Innovations 19 Introduction 02 Platform Architecture 05 Operating Environments 08 Kernel Innovations 10 Cloud Base 14 Enhanced Features 22 Community openEuler 21.09 Technical White Paper 01 Scenario-Specific Innovations 19 Introduction 02 Platform Architecture 05 Operating Environments 08 Kernel Innovations 10 Cloud Base 14 Enhanced Features 22 Community a portal available to global developers, with the goal of building an open, diversified, and architecture-inclusive software ecosystem for all digital infrastructure scenarios. It has a rich history of

● Google percolator and omid ● Internals of TiDB over HBase Features of HBase ● Linear and modular scalability. ● Strictly consistent reads and writes. ● Automatic failover support between RegionServers even more to come. ● Written in Go ○ Faster develop ○ Run fast Why called TiDB ? TiDB Architecture AH. HBase ● First things first ○ Need to build a transactional layer over HBase

Ø Cascades 架构 Ø OLAP性能很棒 Ø https://db.cs.cmu.edu/events/vaccination-2022-orca-a-modular-query-optimizer- architecture-for-vmware-greenplum-venky-raghavan/ Confidential │ ©2022 VMware, Inc. 11 Greenplum的一些概念

buffer overflow)，该漏洞 (CVE-2023-38545) 导致 curl 在 SOCKS5 代理握手过程中溢出基于堆的缓冲区。 Mojo 编程语言开放下载 Mojo 是 Modular AI 公司开发的新编程语言，它结合了 Python 的易用性 以及 C 语言的可移植性和性能，目标是使其成为 AI 研究和生产的理想选 择。 VS Code 实现“浮动窗口”特性 在 2023 TensorFlow 和 PyTorch 和 Hugging Face Transformers 等。 TensorFlow 架构图 (图源：https://www.geeksforgeeks.org/architecture-of- tensorflow/) 52 / 87 LLM 基础设施：编程语言 LLM 的训练和应用通常使用多种编程语言，取决于任务的需求和团 队的偏好。 。它的广泛使用得 益于其简洁的语法、强大的库支持（如 LLM 基础设施：编程语言  2023 年 9 月面向大众开放 创业公司 Modular AI 开 发  结合了 Python 的易用性以及 C 语言的  支持与任意 代码  性能是 Python 的 倍 Mojo�与其他语言性能对比 （图源：https://www.modular.com/max/mojo） 55 / 87 大模型应用现状 2022 年底大模型应用

ShardingSphere‐Proxy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.3 Hybrid Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 4 Running Modes 11 . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Complex primary‐secondary database architecture . . . . . . . . . . . . . . . . . 40 8.3.5 Related References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562 12.7.1 Overall Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562 12.7.2 Encryption Rules

Information Processing Systems 36 (2024). 【6】Karpas, Ehud, et al. "MRKL Systems: A modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoning 问题，中国的计算机科学家给出了答案——科学计算与超级计算机。2007 年，国防科大杨学 军院士带领团队在国际计算机体系结构旗舰会议 ISCA 上发表题为“A 64-bit stream processor architecture for scientific applications”的学术论文[19]，揭示了流处理器架构 可大幅加速科学计算。这项工作不仅为具备大量流处理器的 GPU 开辟了新应用场景，也为超 perceptrons [C], Proceedings HPCA Seventh International Symposium on High-Performance Computer Architecture. IEEE, 2001: 197-206 [2] Mirhoseini, A., Goldie, A., Yazgan, M. et al. A graph placement methodology

instructions for the Debian GNU/Linux 13 system (codename “trixie”), for the 64-bit PC (“amd64”) architecture. It also contains pointers to more information and information on how to make the most of your day-to-day use, but work is continuing. The Hurd is currently only being developed for the i386 architecture, although ports to other architectures will be made once the system becomes more stable. Please requirements beyond the requirements of the Linux kernel and the GNU tool-sets. Therefore, any architecture or platform to which the Linux kernel, libc, gcc, etc. have been ported, and for which a Debian