The graphics processing unit, or GPU was originally designed to accelerate the rendering of 3D graphics. you can use graphics SDK(DirectX, OpenGL) to use GPU for rendering, but it has become one of the most important types of computing technology now, it is designed for parallel processing, the GPU is used in a wide range of applications, including:
Artificial intelligence (AI)Machining learningDeep learningSince the inception of FPGA(field programmable gate array) technology there were actually only two FPGA companies in the market: Xilinx and Altera. Along the years, both companies did a terrific job growing the market and protecting their market share. When Intel acquired Altera, Xilinx was left as the only major FPGA company in the market. Xilinx had ~50% market share while Altera (Intel) has ~37% and Lattice Semiconductor has a 10% market share. In October 2020, AMD has acquired Xilinx. This concludes that the 2 largest CPU vendors have acquired the 2 largest FPGA companies.
Go is a statically typed, compiled programming language designed at Google， Go is syntactically similar to C, but with memory safety, garbage collection, structural typing, concurrency.
Go is influenced by C, but with an emphasis on greater simplicity and safety.
A syntax and environment adopting patterns more common in dynamic languages:
Distinctive approaches to particular problems:
type embedding instead of non-virtual inheritance.by default, produces statically linked native binaries without external dependencies.YUM is the primary package management tool for installing, updating, removing, and managing software packages in Red Hat Enterprise Linux. YUM performs dependency resolution when installing, updating, and removing software packages.
RPM is a popular package management tool in Red Hat Enterprise Linux-based distros. Using RPM, you can install, uninstall, and query individual software packages. Still, it cannot manage dependency resolution like YUM. RPM does provide you useful output, including a list of required packages. An RPM package consists of an archive of files and metadata. Metadata includes helper scripts, file attributes, and information about packages.
Yum is much more useful than rpm as it resolves dependencies of the package and installs them all!, First yum download meta data of packages from enalbed repo, rpm is downloaded only when it’s installing. any another command like search, list only check the meta data.
conf
1 | # show all repos, disabled and enabled |
Name Format of RPM name-version-release.os.arch., example: bash-4.3.2-5.el6.x86_64.rpm
RPM repo
epel: extended packages provided by RedHat, Maintained by Fedora.
1 | # Install |
When scriptlets are called, they will be supplied with an argument. This argument, accessed via $1 (for shell scripts) is the number of packages of this name which will be left on the system when the action completes. So for the common case of install, upgrade, and uninstall we have:
| install | upgrade | uninstall | |
| %pretrans | $1 == 1 | $1 == 2 | (N/A) |
| %pre | $1 == 1 | $1 == 2 | (N/A) |
| %post | $1 == 1 | $1 == 2 | (N/A) |
| %preun | (N/A) | $1 == 1 | $1 == 0 |
| %postun | (N/A) | $1 == 1 | $1 == 0 |
| %posttrans | $1 == 1 | $1 == 1 | (N/A) |
Note: even some scripts in rpm(built-in, no such files after installation) come from .spec, there is NO spec file in xxx.rpm but xxx.src.rpm.
Refer to RPM guideline
It depends on the xx.rpm, as RPM allows to add scripts, hence you can do this there, but most of time, service is not started automatically after installation.
1 | # show all active systemd service |
with noreplace, if it exists, no replace
1 | $ cat libvirtd.spec |
1 | $rpm -qa --last |
You can add %global __os_install_post %{nil} at the top line of xx.spec file.
1 | # remove dependency as well |
rpm may be waiting for a lock to be freed.
1 | $ ps -ef | grep rpm |
libvirt provides lots of tools to manage VM or virtual disk, let’s take a quick look about them.
VM management
virt-install: install VM etcvirsh: start, stop, destroy VM, monitor and collect stats of VM etcvirt-manager: GUI for manage VMVirtual disk management(provided by libguestfs)
guestfish: interactive shell to manage disk(–verbose for debug)guestmount/guestumount: mount/umount disk to host path(–verbose for debug)virt-cat/virt-copy-in/virt-copy-out/virt-format/virt-xxx: commands to manage virtual disk(–berbose for debug)ALL libvirt tools(included libguestfs tools) communicate with libvirt daemon to manage vm or disk by default, but you can direct with qemu if libvirt is not thre
export LIBGUESTFS_BACKEND=direct virt-sysprep
Network block devices (NBD) are used to access remote storage device that does not physically reside in the local machine, for each network block device, it’s mapped with (/dev/nbdx) to client as a local block device, you can do low level operation for this block device, like partition, format with filesystem that NFS can NOT do.
NBD works according to a client/server architecture. You use the server to make a volume available as a network block device from a host, then run the client to connect to it from another host.
Load balancing is a method of distributing IP traffic across a cluster of real servers, providing one or more highly available virtual services. When designing load balanced topologies, it is important to account for the availability of the load balancer itself as well as the real servers behind it.
Keepalived provides frameworks for both load balancing and high availability. The load balancing framework relies on the well-known and widely used Linux Virtual Server (IPVS) kernel module, which provides Layer 4 load balancing. Keepalived implements a set of health checkers to dynamically and adaptively maintain and manage load balanced server pools according to their health.
high-availability is achieved by the Virtual Router Redundancy Protocol (VRRP). VRRP is a fundamental brick for router failover. In addition, Keepalived implements a set of hooks to the VRRP finite state machine providing low-level and high-speed protocol interactions. In order to offer fastest network failure detection, Keepalived implements the Bidirectional Forwarding Detection (BFD) protocol. VRRP state transition can take into account BFD hints to drive fast state transition. Keepalived frameworks can be used independently or all together to provide resilient infrastructures.
In short, Keepalived provides two main functions:
more effort