1.
블록체인과 관련한 기사가 넘칩니다. 흡사 블랙홀 같습니다. 핀테크, 핀테크 하던 때가 엊그제 같은데 핀테크라는 단어보다 블록체인이 훨씬 더 자주 접합니다. 블록체인기술을 ‘판을 갈아엎는(Disruptive) 기술’로 이야기하는 것처럼 자본시장의 대표적인 독점기업이라고 할 수 있는 한국거래소와 예탁결제원도 위기를 느끼는 듯한 발언을 했습니다.
유재훈 예탁결제원 사장은 “블록체인이란 중앙기록관리기관에 의존하지 않더라도 높은 보안과 저렴한 비용으로 데이터를 서로 주고받을 수 있는 것”이라며 “중앙청산소와 중앙매매시스템 등 ‘중앙’이란 이름이 붙어 있는 비즈니스 모델에 급격한 변화가 요구된다”고 설명했다.
(중략)유재훈 사장은 “블록체인이 기존의 예탁결제서비스를 당장 대체하기는 어렵지만, 블록체인이 몰고 올 파괴적인 변화에 대비해 해당 비즈니스를 개발하지 않으면 예탁결제원이 도태될 수도 있다”고 진단했다.
예탁원, 블록체인 확산에 바짝 긴장하는 이유중에서
블록체인기술을 놓고 온통 장미빛 전망을 내놓고 있지만 사실 블록체인기술도 해결할 과제가 산더미처럼 많습니다. 앞서 Hyperledger Project와 블록체인에서 인용하였던 IBM의 자료로 간단히 소개했습니다. 오늘은 구체적으로 살펴보도록 하겠습니다. 앞서 예탁결제원처럼 미국에서 예탁업무를 담당하고 있는 DTCC가 내놓은 보고서는 ‘Limitations of Distributed Ledger Technology’로 일목요연하게 정리하고 있습니다.
DTCC Sharpens Distributed Ledger Focus with White Paper
In assessing the applicability of distributed ledgers to post-trade processing, it is important to understand that the distributed ledger platforms in use today are simply a ledger of transactions that is essentially replicated to all of the cooperating servers. The technology does not have built-in integration with existing systems and supporting infrastructure. It does not simply integrate with user identity management systems or have any master data about legal entities or securities. It does not include supporting workflows, exception processing or any of the extensive preprocessing logic that often accompanies complex matching, allocation and other processes that precede the point at which a transaction is considered complete.
A valuable feature of the distributed ledger platform is that all transactions are considered immutable and never to be modified, cancelled or revoked. The ability for clients to correct/cancel/adjust transactions that were inadver-tently charged or credited to the wrong account is a common occurrence and well managed by today’s financial institutions. Additionally, complex financial transactions often include the ability to reverse the transaction based
on contractual stipulations as a desirable feature. The ability to cancel or reverse a transaction is not supported in today’s distributed ledger platform, and it is not clear today how the platform could evolve to support that.In addition, the technology is lacking in these other areas:
• It does not improve upon existing data retrieval, inquiry, reporting or analytic tools.
• It does not allow data searching in a manner that is even equivalent to modern databases.
• It does not provide high-speed access to data in the same manner as big data technology for purposes of data analytics.
• It does not integrate with modern data management tools.
• It does not address the nonfunctional requirements of most processing systems.For distributed ledgers to reach their full potential, all of these functions, which exist in the current post-trade process and have been constructed over the years by experts and industry professionals, will need to be integrated into this new platform.
Decentralized vs. Centralized Processing
Another important issue to consider in the widespread applicability of distributed ledgers is the trade-offs between decentralized processing and the more traditional computing model of centralized processing. Decentralized processing is, by definition, a shared computing function among members of a community (trusted or not), which requires synchronization and coordination. Some implementations of distributed ledger, such as Bitcoin, use a consensus mechanism to manage coordination, while others use variations such as a lead node mechanism. Regardless, all such designs include steps that add latency to transaction processing. A decentralized design requires significant computing and storage resources because all nodes perform the computations and store the ledger data, which can also result in significantly increased network bandwidth requirements depending on the number of network nodes and the size of each transaction.
In contrast, the use of centralized processing provides a single view of information with a single judgment on a truth, typically on a single machine. The latency of this model is virtually zero. It is certainly possible to take some of the paradigms of the distributed ledger platform, such as improved security, standardized validation rules and verifiable transaction history, and implement them in a centralized system. But this model also requires complete trust in the integrity of the centralized system and in the organization that manages that system.
Global regulatory requirements for data privacy that are different based on geography raise additional challenges for decentralized systems that distribute every transaction to every node. In certain regulatory jurisdictions, the laws protecting an individual’s data privacy restrict the ability to store certain data outside of the regulated region. Several vendors have recently proposed alternative “partitioned” ledgers to address these challenges, but given that all of the current work on distributed ledger technology has been done without regulatory oversight or endorsement, it is still unclear as to the level of regional data containment that will be required.
2.
블록체인의 한계로 가장 많이 언급되는 부분은 Scalabitity입니다. 이와 관련한 문제점을 정리한 글입니다.
The issue of bitcoin scalability and the phrase “blockchain scalability” are often seen in technical discussions of the bitcoin protocol. Will the requirements of recording every bitcoin transaction in the blockchain compromise its security (because fewer users will keep a copy of the whole blockchain) or its ability to handle a great number of transactions (because new blocks on which transactions can be recorded are only produced at limited intervals)? In this article, we’ll explore several meanings of “blockchain scalability” and some high-level technical solutions to the issue.
The three main stumbling blocks to blockchain scalability are:
1) The tendency toward centralization with a growing blockchain: the larger the blockchain grows, the larger the requirements become for storage, bandwidth, and computational power that must be spent by “full nodes” in the network, leading to a risk of much higher centralization if the blockchain becomes large enough that only a few nodes are able to process a block.
2) The bitcoin-specific issue that the blockchain has a built-in hard limit of 1 megabyte per block (about 10 minutes), and removing this limit requires a “hard fork” (ie. backward-incompatible change) to the bitcoin protocol.
3) The high processing fees currently paid for bitcoin transactions, and the potential for those fees to increase as the network grows. We won’t discuss this too much, but see here for more detail.
Blockchain scalability중에서
앞서와 비슷한 이야기지만 금융서비스에 블록체인을 적용하려면 비용과 속도를 해결하여야 한다고 주장을 합니다.
4) Cost. Some would argue that the blockchain technology underpinning bitcoin is free. I would disagree with that view, given that the computational effort needed to create and maintain the Bitcoin blockchain is immense. In fact, blockchain technology could be perceived as computationally intensive and wasteful. The cost of compute power (or electricity) needed to process a single transaction, may make the transaction cost too high to be cost effective for most financial services use cases. Within a cost conscious commercial context, the industry may therefore need to implement blockchain technology in ways that minimise the computational effort needed and the cost per transaction.
5) Speed. Perhaps one of the most limiting factors of blockchain technology, at least in the way it has been implemented by Bitcoin, is that it can be very slow. Bitcoin transactions can take hours to be confirmed, which would rule out quite a few financial use cases. However, there are ways in which this limitation can be addressed through alternative implementations.
Blockchain and Financial Services중에서
3.
가능성과 한계가 공존하고 있는 블록체인. 모든 것을 완벽히 해결하는 것이 나올 수 있을지 의문입니다. 어느 선에서 타협을 하겠지만 역시나 관건은 각 나라가 만들어놓은 규제입니다. Blockchain: Need of future for Financial Industry는 규제와 제도와 연결하여 한계를 논하고 있습니다. 규제와 맞추려다가 블록체인의 가능성이라고 했던 본질을 바꿔야 하는 일도 발생할 수 있습니다. 갈 길이 멉니다.
Below are some of the Limitations and Risk of Blockchain method:
Compliance Check
Compliance check would be full responsibility of the Message Initiating Party.
Compliance check is not possible with current frame work of transmission of messageOngoing development
Blockchain software is still in pre-matured and incomplete features with an active development
New tools, features, and services are required to make Blockchain more secure and accessible to the massesTransaction Recall
Blockchain method of processing do not have any facility of transaction recall or cancellation request processingScarcity of Resources
Resources who are knowledgeable in Blockchain processing as Developer or testerRegulatory Limitations
There are not clear regulations currently for Blockchain transactions
