Summary of several memory data and several memory data

Summary of several memory data and several memory data

MemSQL

Principle: combines unlocked data structures with real-time compilers to handle large workloads, lock-free hashtable and lock-free skiplists (a hierarchical linked list structure) are implemented in the memory to quickly access data randomly.

Persistence: Although MemSQL stores data in the memory, it can persist data to the disk by writing logs and snapshots (similar to checkpoint ).

Replication: MemSQL currently supports master-slave replication. It supports local replication protocol to transfer transaction logs to slave.

Distributed architecture: working based on the concepts of aggregators and leaf nodes, a leaf node is a MemSQL database. The aggregators are responsible for decomposing and querying the relevant leaf nodes, and aggregating the results back to the client.

GemFireThe technical principles of the memory-based distributed cluster system and distributed memory data platform are shown in: the memory of several X86 servers is centralized through the virtualization technology of the cloud computing platform, A memory resource pool consisting of up to dozens of terabytes of data can be loaded into the memory for memory computing. The computing process itself does not need to read or write data to the disk, but regularly writes data to the disk in synchronous or asynchronous mode. GemFire stores multiple copies of data in a distributed cluster. If one machine fails and other machines have backup data, you do not have to worry about data loss and have disk data as a backup. GemFire supports the persistence of memory data to various traditional relational databases, Hadoop libraries, and other file systems. Earlier than 12306, it adopted the Unix minicomputer architecture and transformed it into a Linux/X86 server cluster architecture using GemFire technology, it means that it spans three generations. From minicomputers to large memory X86 server clusters, this not only improves the performance by an order of magnitude, but also reduces the cost.

Neo4jIs an embedded, disk-based Java persistence engine that supports complete transactions. It stores data in Graphs (networks) rather than tables. Neo4j provides large-scale scalability. It can process billions of nodes, links, and attributes on one machine, and can be extended to multiple machines for parallel operation. Compared with relational databases, graph databases are good at processing a large amount of complex, interconnected, and low-structured data. These data changes rapidly and requires frequent queries-in relational databases, these queries cause a large number of table connections, which may cause performance problems. Neo4j focuses on solving the performance degradation problem of traditional RDBMS with a large number of connections during query. By modeling data around a graph, Neo4j traverses nodes and edges at the same speed, and its traversal speed has nothing to do with the data volume that forms the graph. In addition, Neo4j provides very fast graph algorithms, recommendation systems, and OLAP-style analysis, which cannot be implemented in the current RDBMS system.

VOLTDBIt is an open-source OLTP database running on clusters and the only memory solution for intelligent and fast data. It provides memory performance and NoSQL scalability, full stream capability and continuity of traditional relational databases provide reliability and fault tolerance in the secret room and cloud environment. Data Tables are divided into multiple servers in the cluster. They are compatible with SQL and ACID and can be expanded online. Data Consistency is ensured through parallel single-thread processing, the traditional database locks and resource management overhead are avoided. However, because VOLTDB data analysis is based on SNA, the data distribution policy is based on hash, the cluster Scale is determined in advance. New machines need to stop services and then re-distribute data. In addition, after data hash is dispersed, data continuity is disrupted, in this data structure, the machines used to serve this table are required for range query, and the query efficiency is very low.

SAP HANAHana is mainly used for primary data analysis. hana can extract other types of data and perform quick analysis on hana, the main difference between sap hana and traditional analysis models is that they discard any material, that is, all models are completely virtual and are calculated based on the basic operation data, the model can be easily modified. It uses a data dictionary method to compress data and minimize data transmission. It distributes large amounts of data and computing workload to different processors and shares a set of data between different servers, A single server DOWN will not affect any computing, and the cost is extremely expensive, tens of millions.

What kind of memory stick? How to identify the size? Model?

When saving a computer, we generally only pay attention to the memory capacity and memory performance indicators, such as DDR266 DDR333. But do you know that the same memory has different performance and stability depending on its brand, factory time, and batch number, this article will focus on the differences in performance and stability between different brands of memory and the performance differences between memory of the same brand but different batches. In addition, this article will focus on the memory counterfeiting and sales methods. The Identification Method of genuine memory and Shenzhen oil bar will be thoroughly disclosed to you. (What is oil strip? We will not sell soy milk here. We believe that we will teach you how to use fried dough sticks)
Considering that SDRAM has reached the end of its life and is about to be completely removed from the market, the RDRAM application is not widely used and the DDRII memory has not yet been produced. Therefore, this article will only involve mainstream DDR memory.
Let's take a look at the basic knowledge of memory and summarize it into one sentence: What is memory?
The Memory is the Random storage (Random Access Memory ). RAM is divided into two categories: Static random memory, Static RAM (SRAM) and Dynamic random memory, Dynamic RAM (DRAM). We often say that "system memory" refers to the latter, DRAM.
SRAM is an important memory. It is widely used in various fields. The speed of SRAM is very fast. Data integrity can be maintained during fast reading and refresh, that is, data is not lost. The internal structure of SRAM stores data in the form of a bistability circuit. To achieve this structure, the structure of the SRAM circuit is very complex. A large number of transistors are often used to construct registers to retain data. Using a large number of transistors requires a large amount of silicon, which increases the chip area and virtually increases the manufacturing cost. Creating an SRAM of the same capacity is much more expensive than DRAM. Therefore, the SRAM can only be used for the first-level cache inside the CPU and the built-in second-level cache on the PC platform. What we call "system memory" should be DRAM. Because of the high cost of SRAM, its development is severely restricted. Currently, only a small number of network servers and routers use it.
DRAM is more widely used than SRAM. DRAM structure is much simpler than SRAM. It only consists of one MOS tube and one capacitor. Therefore, DRAM is superior to SRAM in terms of integration, production cost and volume. At present, with the continuous development of PC, we have more and more requirements for system memory. With the launch of WindowXP, the software is more dependent on memory: in Windows XP, the Professional Edition requires at least MB of memory. In actual use, MB can ensure the normal operation of the system. Therefore, with the development of PC, the memory capacity will continue to expand, and the speed will continue to increase.
Now let's talk about the memory speed problem. The memory speed is determined by the Front-End bus of the CPU. For example, if you use P4A, you do not need to use DDR400 to meet the bandwidth requirement of 3.2gb because the front-end bus of P4A is 400 MHZ, ordinary DDR266 can provide 2. 1 GB of bandwidth. This memory is suitable for low-frequency and low-end configurations such as ATHLONXP and Poison Dragon, which are no longer in the mainstream market. DDR333 can provide GB memory bandwidth for AMD166MHZ outband Barton processors. DDR400 memory and DDR433 DDR450 memory will be able to provide more than GB of memory bandwidth, mainly used on INTEL's high-end P4C P4E and ATHLON64. Of course, the above is only to meet the minimum requirements of the hardware system. Due to the improvement of the p4 Front-End bus, the bottleneck problem between memory and CPU is already very serious, and the new internal bypass technology will surely emerge. for example, we will briefly introduce the dual-channel DDR and future DDR memory. of course, if you have a Poison Dragon 1.6G, ATHLONXP1800 + (b0 1.5V low-voltage version), P41.8A, AthlonXP2000 +, p42.4C button 2500 + CPU with the best super product potential, so I recommend you purchase a high-performance memory, such as Athlon2000 + ...... remaining full text>

Which memory types are available?

Frequently seen memory usage is as follows: EDO, SD, DDR, DDR2, and DDR3.

In fact, there are many memory types.

Based on the components, RAM memory is divided into the following 18 types:

01. DRAM (Dynamic RAM, Dynamic Random Access Memory ):

This is the most common RAM. An electronic tube and a capacitor form a storage unit. DRAM stores each memory bit as a charge in the storage unit, and uses the charge and discharge of the capacitor for storage, however, because the capacitor itself has a leakage problem, it must be refreshed every several microseconds. Otherwise, data will be lost. The access time is the same as the discharge time, which is about 2 ~ 4 ms. Because the cost is relatively low, it is usually used as the primary storage in the computer.

02. SRAM (Static RAM, Static random access memory)

Static means that the data in the memory can be stored in it without being accessed at any time. Each six tubes form a single storage unit. Because there is no capacitor, it can operate normally without constant charging. Therefore, it can process faster and more stably than the general dynamic random Processing Memory, it is often used for high-speed caching.

03. VRAM (Video RAM, Video memory)

Its main function is to output video data from a video card to a digital-to-analog converter, effectively reducing the workload of the graphics display chip. It adopts a dual-data port design. One of the Data ports is a parallel data output entry, and the other is a serial data output port. It is mostly used in high-end memory of graphics cards.

04. fpm dram (Fast Page Mode DRAM, Fast Page switching Mode Dynamic Random Access Memory)

Improved DRAM, most of which are 72Pin or 30Pin modules. When traditional DRAM accesses a BIT of data, it must send the row address and column address each time to read and write data. After frm dram triggers the row address, if the address required by the CPU is in the same row, the column address can be output consecutively without having to enter the travel address. Since the general program and data are arranged consecutively in the memory, in this case, the required data can be obtained after the row address is output consecutively. FPM splits the memory into many Pages, ranging from B to several KB. When reading data in a continuous area, you can use the Quick page switching mode to directly read the data on each page, thus greatly improving the reading speed. 96 years ago, fpm dram was widely used in the early stages of the 486 and PENTIUM eras.

05. edo dram (Extended Data Out DRAM, Extended Data output dynamic random access memory)

This is a kind of memory that emerged after FPM, generally 72Pin and 168Pin modules. It does not need to output the row address and column address when accessing each BIT of data like fpm dram and make it stable for a period of time before it can read and write valid data, the next BIT address can be output only after this read/write operation is complete. Therefore, it can greatly shorten the waiting time for the output address, and its access speed is generally about 15% faster than that in FPM mode. It is generally used in the standard memory of the Pentium motherboard below the middle level. Later versions of the 486 system began to support edo dram, and later versions started to run edo dram ..

06. bedo dram (Burst Extended Data Out DRAM, Burst Extended Data output dynamic random access memory)

This is an improved edo dram, proposed by micron, which adds an address counter on the chip to track the next address. It is a burst reading method, that is, when a data address is sent out, each of the three remaining data can be read in only one cycle. Therefore, multiple groups of data can be accessed at a time, faster than edo dram. However, the number of mainboards supporting bedo dram memory is very small, and only a few of them provide support (such as via apollo VP2. therefore, DRAM was quickly replaced.

07. MDRAM (Multi-Bank DRAM, Multi-slot Dynamic Random Access Memory)

A memory specification proposed by MoSys, Which is internally divided into... the remaining full text>